uefi 5.1 x370 taichi EASTER egg surprise
Printed From: ASRock.com
Category: Technical Support
Forum Name: AMD Motherboards
Forum Description: Question about ASRock AMD motherboards
URL: https://forum.asrock.com/forum_posts.asp?TID=10677
Printed Date: 03 Nov 2024 at 9:22pm Software Version: Web Wiz Forums 12.04 - http://www.webwizforums.com
Topic: uefi 5.1 x370 taichi EASTER egg surprise
Posted By: cx5
Subject: uefi 5.1 x370 taichi EASTER egg surprise
Date Posted: 14 Jan 2019 at 10:34am
My ram is Corsair CMK32GX4M2B3000C15 = 16GB x 2 (dual rank dual side) 3000Mhz C15, Hynix MFR detected by Thaiphoon 1.4 or earlier. Can be seen my older posts.
But uefi 5.1 has a new feature and detected it as 3200Mhz C16 with part number CMK32GX4M2B3200C16 and version 4.31, Hohohoho I was so darn happy :) imagine that !!!!
then Thaiphoon 1.4 no longer reads out and must upgrade to 1.5 but then the MFR turned into AFR - WTH ........ but the timings nanoseconds are the same..... 3000Mhz also same etc.
and needless to say I tried booting 3200Mhz and failed...... 3 long beeps, but with experience from previous experiments + powerful Dram Calculator by 1usmus, I quickly up the voltage to Dram 1.5v, soc 1.2v and changed 3 timings to follow calculator 1.4.0.1 and huala boot and game and no crash 3200Mhz C16181818...
------------- x370taichi@3.50gen4 AX750 Optane900P SN770 3800x PosdonZro4 BarrowPWM-17w>420x140x28 XFX SWFT319 6900XT MSI RX560 LP to get Fluid Motion BlueskyFRC 75hz
|
Replies:
Posted By: gizmic
Date Posted: 14 Jan 2019 at 12:07pm
its been like this since 2 beta bios back it happens to all ram regardless what you use lets say its the safest ryzen ram profile for 3200 it also help people using Oem rams that doesnt have xmps
so now we can load our kits xmp and a cl 16 3200 xmp
-------------
|
Posted By: cx5
Date Posted: 14 Jan 2019 at 1:24pm
hahahahaha, I look to stupid now, I was on 4.81 beta but didn't see it...h hahahaah
then I should give it a try its full timing pacakage 3200 and see what happens with added voltages only, Thanks.
------------- x370taichi@3.50gen4 AX750 Optane900P SN770 3800x PosdonZro4 BarrowPWM-17w>420x140x28 XFX SWFT319 6900XT MSI RX560 LP to get Fluid Motion BlueskyFRC 75hz
|
Posted By: gizmic
Date Posted: 14 Jan 2019 at 4:53pm
haha well since the pinnacle 1006 i can boot 3466 cl4@1.385v while i was very unstable at 3333 very relaxed timings. You may want to give it another go.
there is another beta bios that has weird ram straps (3533&3666) but i think its meant for bclk oc i'll test them at the end of a month as i'm going to vacation on friday.
-------------
|
Posted By: cx5
Date Posted: 15 Jan 2019 at 1:14pm
gizmic wrote:
i'm going to vacation on friday. |
Happy holiday !!!
I tried default timing from the easter egg, but no luck, in fact last night all 3200Mhz bad, got to push up voltage to 1.52 and must use 1usmus calculator timing for my sticks, hence gave up default easter egg.
Still, no problem with my previous 2933 is good enough.
------------- x370taichi@3.50gen4 AX750 Optane900P SN770 3800x PosdonZro4 BarrowPWM-17w>420x140x28 XFX SWFT319 6900XT MSI RX560 LP to get Fluid Motion BlueskyFRC 75hz
|
Posted By: J Z
Date Posted: 15 Jan 2019 at 5:05pm
Hello,
the CMK32GX4M2B3200C16 v4.31 is a special features only for version 4.31 with Samsung chips !!!
Since no XMP is read out wrong, it is a feature!
------------- Kind Regards, JZ
https://shop.JZelectronic.de - Der Shop mit ausgesuchter ASRock Profi Hardware
https://www.facebook.com/asrock.de
|
Posted By: gizmic
Date Posted: 16 Jan 2019 at 10:54am
change to samsung la cx5 make things complicated
-------------
|
Posted By: cx5
Date Posted: 23 Jan 2019 at 3:54pm
J Z wrote:
Hello,
the CMK32GX4M2B3200C16 v4.31 is a special features only for version 4.31 with Samsung chips !!!
Since no XMP is read out wrong, it is a feature! |
Oh I see, "it's a feature", now I know and althought I don't have samsung chip to use it, but I like it, for a short moment gave me extra hope, that's happy too :) haahahah
------------- x370taichi@3.50gen4 AX750 Optane900P SN770 3800x PosdonZro4 BarrowPWM-17w>420x140x28 XFX SWFT319 6900XT MSI RX560 LP to get Fluid Motion BlueskyFRC 75hz
|
Posted By: cx5
Date Posted: 23 Jan 2019 at 3:57pm
gizmic wrote:
change to samsung la cx5 make things complicated |
haih...... gotto sell off my sticks first... but 16GB x2 is hard to find buyer ...... :(
at least it runs 2933 really well at C14,15,15,15 not bad lah..... somemore at lower voltages 1.33 instead of 1.35
If you have friend interested, please let me know, then next time I buy smallest nanometer dram, my stick is 25nm hence lost big time to samsung's 21nm??? 20nm???
heard 14nm is coming out for every other brand now.....
------------- x370taichi@3.50gen4 AX750 Optane900P SN770 3800x PosdonZro4 BarrowPWM-17w>420x140x28 XFX SWFT319 6900XT MSI RX560 LP to get Fluid Motion BlueskyFRC 75hz
|
Posted By: paologab
Date Posted: 23 Jan 2019 at 6:03pm
hi me too I'm quiet buffled with the last bios.
I'm running my dram @3400 (Kingston with Hinix MFR rated @3200) with the timings of XMP@3000.
The only changes made are auto OC voltage, offset voltage -0,87, both LLC @3.
Many full memtest (made @3066, @3200, @3333, @3400) did show zero errors.
------------- AMD Ryzen 7 2700X w/Wraith prism x470 Taichi bios 3.50 16gb HX432C16PB3K2/16 (hynix H5AN8G8NCJR-UHC) M.2 970 EVO, SATA SSD, 2 SATA HDD AMD Radeon RX 580 8G OC Linux Kernel 5.4 Manjaro Stable XFCE
|
Posted By: cx5
Date Posted: 24 Jan 2019 at 5:01pm
paologab wrote:
hi me too I'm quiet buffled with the last bios.
I'm running my dram @3400 (Kingston with Hinix MFR rated @3200) with the timings of XMP@3000.
The only changes made are auto OC voltage, offset voltage -0,87, both LLC @3.
Many full memtest (made @3066, @3200, @3333, @3400) did show zero errors.
|
WoW, you so luck !!!!! Hynix ram so good ???? what part number? may be is newest 14nm type ???
------------- x370taichi@3.50gen4 AX750 Optane900P SN770 3800x PosdonZro4 BarrowPWM-17w>420x140x28 XFX SWFT319 6900XT MSI RX560 LP to get Fluid Motion BlueskyFRC 75hz
|
Posted By: paologab
Date Posted: 24 Jan 2019 at 6:29pm
I don't know. I don't think, is on the QVL, Hyperx PredatorHX432C16PB3K2/16. Btw @3466 boots and works ok, but memtest gives errors. I'll stay @3400 and fine tune the offset.
------------- AMD Ryzen 7 2700X w/Wraith prism x470 Taichi bios 3.50 16gb HX432C16PB3K2/16 (hynix H5AN8G8NCJR-UHC) M.2 970 EVO, SATA SSD, 2 SATA HDD AMD Radeon RX 580 8G OC Linux Kernel 5.4 Manjaro Stable XFCE
|
Posted By: paologab
Date Posted: 25 Jan 2019 at 1:33am
strange enough @3400 a Windows 10 WinPE ISO loads all the stuffs and then freeze with the window logo.
------------- AMD Ryzen 7 2700X w/Wraith prism x470 Taichi bios 3.50 16gb HX432C16PB3K2/16 (hynix H5AN8G8NCJR-UHC) M.2 970 EVO, SATA SSD, 2 SATA HDD AMD Radeon RX 580 8G OC Linux Kernel 5.4 Manjaro Stable XFCE
|
Posted By: cx5
Date Posted: 25 Jan 2019 at 4:59pm
I never do memtest, no patience, I only use game to test, if no crash then good enough, hahhahhhaa lazy way.
------------- x370taichi@3.50gen4 AX750 Optane900P SN770 3800x PosdonZro4 BarrowPWM-17w>420x140x28 XFX SWFT319 6900XT MSI RX560 LP to get Fluid Motion BlueskyFRC 75hz
|
Posted By: paologab
Date Posted: 26 Jan 2019 at 12:05am
it's better as memtest doesn't push the pc to his limit... two pigeons with a stone!
------------- AMD Ryzen 7 2700X w/Wraith prism x470 Taichi bios 3.50 16gb HX432C16PB3K2/16 (hynix H5AN8G8NCJR-UHC) M.2 970 EVO, SATA SSD, 2 SATA HDD AMD Radeon RX 580 8G OC Linux Kernel 5.4 Manjaro Stable XFCE
|
Posted By: xhue
Date Posted: 26 Jan 2019 at 12:22am
Real memtest86 is designed specifically to test with word patterns, which are quite exhaustive. The goal is to use data words that has the highest probability of causing instability or catching other issues like bit flipping, etc.
While one shouldn't dismiss easily the full power of memtest86, it's always wise to test with other apps as well.
Also, note that many programs call themselves memtest while in practice are quite far from the real one.
|
Posted By: paologab
Date Posted: 26 Jan 2019 at 12:48am
hi xhue
[ot]
thanks to point that out. It was one of my doubt whether MemTest86 (the one licensed for free by PassMark(TM)) is thaft enough.
On my pc I installed linux from scratch in mixed UEFI-legacy mode, so I could use memtest86+ (should be the MemTest86+ also passed to PassMark). This is created as boot option to test the ram by almost any distribution.
Then I reverted the boot to UEFI only so memtest86+ didn't work anymore and I had to switch to MemTest86 that can be executed at UEFI boot.
Compared to MemTest86, memtest86+ cause much more heat. I mean after a full pattern test made with memtest86+ the chassis was much more warm than it is after 4 passes of full MemTest86. Both ran on all cores, in parallel, with all tests.
[/ot]
------------- AMD Ryzen 7 2700X w/Wraith prism x470 Taichi bios 3.50 16gb HX432C16PB3K2/16 (hynix H5AN8G8NCJR-UHC) M.2 970 EVO, SATA SSD, 2 SATA HDD AMD Radeon RX 580 8G OC Linux Kernel 5.4 Manjaro Stable XFCE
|
|