|
AGESA 1.0.0.6b |
Post Reply 
|
Page <1 56789 12> |
| Author | ||
Denroth 
Newbie 
Joined: 02 Apr 2017 Status: Offline Points: 46 |
 Post Options
 Thanks(0)
 Quote Reply
 Posted: 16 Sep 2017 at 10:05pm |
|
Thanks man <3 with the new bios I can overclock ram at 3066mhz :D (stable) https://i.imgur.com/qggKdLD.png  Edited by Denroth - 16 Sep 2017 at 10:06pm |
||
|
Ryzen R7 1700 3.8ghz 1.3v| Msi R9 390 8GB DDR5| Ram GSkill Ripjaws V 3200mhz b-die F4-3200C15D-16GVR(3066mhz)
|
||
 |
||
|
GabyGancia
Newbie
Joined: 17 Sep 2017 Status: Offline Points: 1 |
Post Options
Thanks(0)
Quote Reply
Posted: 17 Sep 2017 at 7:44am |
|
|
Ryzen R7 1700 @3.7Ghz
Asrock X370 Gaming K4 Bios 3.00 Corsair 2x8Gb @2666Mhz CMR16GX4M2C3000C15 |
||
|
||
|
lowdog
Newbie
Joined: 16 Apr 2017 Location: Australia Status: Offline Points: 194 |
Post Options
Thanks(0)
Quote Reply
Posted: 19 Sep 2017 at 6:14pm |
|
I'm on the X370 Pro Fat Gaming and bios 3.10 with the 1.0.0.6b Agesa does indeed have a vcore problem. With the same P state setting I did with bios 3.00 which gave me a solid stable vcore of 1.344/1.36 load with Level 5 LLC this bios will give the same vcore with the same settings.... BUT!!!...it gradually creeps up over time till it's sitting at 1.4V almost constantly, even when idle and hardly down clocks.....temps are pushing 50C when the comp is just idleing. This bios has a SEVERE vcore issue that need reporting to Asrock ASAP before cpu's start frying. Going to flash back to 1.0.0.6a bios 3.00 till this overvolting vcore creep up is sorted in a new bios. Edited by lowdog - 20 Sep 2017 at 4:59am |
||
|
X399 Fat Pro Gaming bios 3.10 - TR4 1900X - 64GB G-SKILL TridentZ F4-3200C14Q-64GTZ @ 3133MHz - Vega 64 AIO with EK block - WC Custom loop - 1500W Silverstone PSU - yay
|
||
|
||
|
datonyb
Senior Member 
Joined: 11 Apr 2017 Location: London U.K. Status: Offline Points: 3139 |
Post Options
Thanks(0)
Quote Reply
Posted: 20 Sep 2017 at 1:26am |
|
|
[url=https://valid.x86.fr/jpg250]
 [/url]3800X, powercolor reddevil vega64, gskill tridentz3866, taichix370, evga750watt gold |
||
|
||
|
seanpatrick
Newbie
Joined: 28 Apr 2017 Location: Toronto Status: Offline Points: 8 |
Post Options
Thanks(0)
Quote Reply
Posted: 20 Sep 2017 at 2:47am |
|
|
This is a problem with the BIOS and p-state overclocking, perhaps with voltage offsets - it's not attributable to LLC adjustments in this context.
|
||
|
||
|
lowdog
Newbie
Joined: 16 Apr 2017 Location: Australia Status: Offline Points: 194 |
Post Options
Thanks(0)
Quote Reply
Posted: 20 Sep 2017 at 4:57am |
|
As I mentioned these are the exact same settings I had in bios 3.0 with Agesa 1.0.0.6a and it did not exhibit the same behaviour that is occurring with bios 3.10 with Agesa 1.0.0.6b......the problem lies in the new bios. Edited by lowdog - 20 Sep 2017 at 4:58am |
||
|
X399 Fat Pro Gaming bios 3.10 - TR4 1900X - 64GB G-SKILL TridentZ F4-3200C14Q-64GTZ @ 3133MHz - Vega 64 AIO with EK block - WC Custom loop - 1500W Silverstone PSU - yay
|
||
|
||
|
SternRabbit
Newbie
Joined: 20 Sep 2017 Status: Offline Points: 7 |
Post Options
Thanks(0)
Quote Reply
Posted: 20 Sep 2017 at 9:18pm |
|
|
I agree with lowdog, something indeed is wrong.
Same setup here with pstates but volts could be up to 0.05v higher than before or lower. My lowest voltage is 0.384v vCore and highest up to 1.41v vCore Previously it was 0.880v vCore and 1.36v vCore. If i have 1.36v i have a low of 0.384v If i have 1.41v i have a low of 0.880v Changes with every boot CPU LLC1 SoC LLC3 Edited by SternRabbit - 20 Sep 2017 at 9:36pm |
||
|
||
|
nangu
Newbie
Joined: 06 Jul 2017 Status: Offline Points: 120 |
Post Options
Thanks(0)
Quote Reply
Posted: 20 Sep 2017 at 11:30pm |
|
|
Was a little bored yesterday and decided to take another shoot at memory OC.
Tried 3066, managed to get it past boot and stress test it on Windows. After two hours of stress testing while watching a stream without errors, decided to look at some benchmarks to compare with my previous results @2933. Rebooted to UEFI, adjusted fan curves a bit, save, restart, and bam!! Dr Debug's F9 festival. I couldn't boot anymore with memory @3066 :-( I think AMD and motherboard's vendors need to address this cold boot / memory training problems. I find hard to believe a setting which is stable at stress testing, never sticks at reboot or cold boot. Tha same applies at memory's voltage at boot. I can run perfectly stable @2933 at my memory rated voltage of 1.35v, but It need 1.375v to avoid F9 fest at cold boot. Hint: We need memory boot voltage setting in UEFI as ASUS boards. Also, going from @2933 to @3066 needs a lot of voltages increase (when it posts if you're lucky enough). VDram from 1.375v to 1.40v, SoC from 0.92v to 1.1v. I don't know where the culpruit lies: It's a weak IMC controller on the CPU? It's a memory automatic settings and/or training issues? It's a poor DDR4 design on motherboard? I know AMD's specs lists @2666 as max memory supported btw, so @2933 I can reach stable is a bonus actually, but I think there is a problem somewhere anyway, because a stress test stable RAM @3066 on Windows which doesn't past training 9 of 10 times, it's an AGESA and/or UEFI design problem in my books. Thanks for reading, and sorry for my little rant :-)
|
||
|
R7 1700 @3.90 1.25v - GSkill TridentZ 3200c16 Hynix MFR @2933 14-16-16-32 - Fatal1ty Gaming X370 K4 - Gigabyte GTX 1070 G1 - WD 256 Black M2 Nvme as Windows 10 boot drive - EVGA Gold 650W
|
||
|
||
|
wardog
Moderator Group 
Joined: 15 Jul 2015 Status: Offline Points: 6447 |
Post Options
Thanks(0)
Quote Reply
Posted: 21 Sep 2017 at 1:02am |
|
|
nangu, was that with Memory Training :Disabled ?
Hint: Or, my experiences here........ Start with SoC at 1.1, VDram at ? called for, VTT_DDR runs half of VDram and I'll and a pinch, like, if VDram at 1.375 I'll make VTTT_DDR 0.69/0.695/0.700 SoC you can back down once your mem begins responding. You'll know when it "begins responding" as it's like being perched on the fence. A little this way, a little that way, and all will still be stable. Versus pushing it all uphill. |
||
|
||
|
nangu
Newbie
Joined: 06 Jul 2017 Status: Offline Points: 120 |
Post Options
Thanks(0)
Quote Reply
Posted: 21 Sep 2017 at 3:42am |
|
|
Hi wardog,
Thanks for your suggestions. No, it was with memory training enabled. On my system I found I had to leave it enabled to not get one or two F9 errors on cold boot @2933, but I will try disabling that option @3066 and see how it does. VTT_DDR I set always at half VDram. My main problem now is to get a config @3066 which passes post. Once I get it nailed (if I ever can), yes, as you said I could start lowering SoC volts and testing stability. My main gripe right now is how frustrating it is when you pass two hours of stress testing only to discover your "somewhat OK" settings do not pass a reboot and/or a cold post never again :-(( I'm having dreams at night with that "F9" Dr Debug code :-))) Thank you very much anyway, I'll wait for the new AGESA (if AMD is kind enough to release it some day) to try again. I'm happy with my system right now, but you know, we always want more and more, and tinkering with this AMD platform had become both a nightmare and a very good learning excercise ;-) Cheers,
|
||
|
R7 1700 @3.90 1.25v - GSkill TridentZ 3200c16 Hynix MFR @2933 14-16-16-32 - Fatal1ty Gaming X370 K4 - Gigabyte GTX 1070 G1 - WD 256 Black M2 Nvme as Windows 10 boot drive - EVGA Gold 650W
|
||
|
||
|
Post Reply
|
Page <1 56789 12> |
Tweet
|
| Forum Jump | Forum Permissions You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |
Topic Options
wardog wrote: