A300 DeskMini Boot Problems

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   Dear Amiga1000,

Please help to check when USB SSD connected, the BIOS boot option priority#1 is the Windows Boot Manager.

Thank you!
Yours truly,
ASRock TSD

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
ASRock_TSD Members Profile Send Private Message Find Members Posts Add to Buddy List ASRock_Official Joined: 20 Mar 2015 Status: Offline Points: 8787	Post Options Post Reply Quote ASRock_TSD Report Post Thanks(0) Quote Reply Posted: 16 Jul 2020 at 2:19pm
	Dear Amiga1000, Please help to check when USB SSD connected, the BIOS boot option priority#1 is the Windows Boot Manager. Thank you! Yours truly, ASRock TSD

J Z Members Profile Send Private Message Find Members Posts Add to Buddy List Groupie Joined: 09 Sep 2016 Location: Germany Status: Offline Points: 976	Post Options Post Reply Quote J Z Report Post Thanks(0) Quote Reply Posted: 16 Jul 2020 at 3:25pm
	Hello, I can confirm that starting with the L3.60G and L3.60H there are boot problems with USB. Sometimes it is sufficient if there is only a USB stick or a USB keyboard / mouse, after which the system simply crashes. No problems with interer SSD I always test the Deskmini A300 with my MasterSSD under USB and so far it has worked perfectly for all BIOS versions. Only with the AGESA 1004 a USB boot with my USB devices is not possible. https://www.lc-power.com/produkte/hdd-ssd-m2-gehaeuse/635cm25/lc-25bub3/ But it's my turn with Sean and I'm looking for a permanent solution with AGESA 1004
	Kind Regards, JZ https://shop.JZelectronic.de - Der Shop mit ausgesuchter ASRock Profi Hardware https://www.facebook.com/asrock.de

Amiga1000 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 15 Jul 2020 Location: Germany Status: Offline Points: 4	Post Options Post Reply Quote Amiga1000 Report Post Thanks(0) Quote Reply Posted: 17 Jul 2020 at 12:03am
	Thanks for the quick answer! Yes, I checked all BIOS options and they are correct - as far as I can see. I found a workaround for the problem: In the boot options I switch "Fast Boot" to "Ultra fast". The external USB drives will then be ignored and the boot process works fine. But maybe you can inform R&D about this bug in BIOS 3.60H. Some more bugs: - DeskMini 300 doesn't run stable with "HyperX Impact DDR4 HX429S17IB2K2" if I run it at the recommended 2933 MHz. The PC switches off under Windows 10. If I set the RAM to 2866 MHz it works fine. The HyperX RAM is on your memory support list, so it should work. But it doesn't. - the "ON"-switch of the Deskmini 300 doesn't work sometimes. I have to pull the power cable and reconnect it, then I have a new chance. - my external Samsung T5 SSD only runs at 40 MB/sec at the Deskmini USB-C port. But since it is USB 3.1 Gen1 it should run at 400-500 MB/sec. This is very disappointing. Thanks for some hints. Torsten

Hammerfest Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 05 Jun 2020 Status: Offline Points: 21	Post Options Post Reply Quote Hammerfest Report Post Thanks(0) Quote Reply Posted: 17 Jul 2020 at 8:39am
	Change log for AGESA 1005 Rollup of 1004a, ab, abb, abba patches into a single release Fixed a PCIe® lane configuration issue on the AMD Ryzen??3 PRO 2100GE Resolved an intermittent virtual memory error with Realtek onboard LAN Improved POST with select Micron DDR4-3200 memory ICs Optimized PCIe® firmware to improve stability and interoperability Considering the A300 doesn't actually have a chipset and is using those direct PCIe lanes to the chip, this could be relevant. Sadly, there is no public changelog for 1.0.0.6 which is probably final or near final beyond any major bug fix's from AMD, so cant tell if theres anything else which might impact direct PCIe lanes. I know your issue is booting from the SSD, what OS? "- my external Samsung T5 SSD only runs at 40 MB/sec at the Deskmini USB-C port. But since it is USB 3.1 Gen1 it should run at 400-500 MB/sec. This is very disappointing." I have a WindowsToGo boot drive and due to drivers and the way windows works, this is actually a thing on windows with some USB implementations, I will dust it off and give it a go, is this through the Type-C port only, or have you tried with the Type-A port on the front (same grouping)

Amiga1000 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 15 Jul 2020 Location: Germany Status: Offline Points: 4	Post Options Post Reply Quote Amiga1000 Report Post Thanks(0) Quote Reply Posted: 17 Jul 2020 at 3:30pm
	Thanks for the input regarding the direct PCIe lanes, which I wasn't aware of. I hope there will be updates for new AGESA versions in future and ASRock will provide them to us. My Deskmini boots from an internal M.2 SSD (Samsung 907 Pro), and the low speed affects only the external Samsung T5 drive. Yes, I tried the USB-A port and this is pretty fast with 380 MB/s. I wonder why USB-C is so much slower - I don?�t have an explanation. I already tried two different cables that are suitable for 10 Gbit/s.

Hammerfest Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 05 Jun 2020 Status: Offline Points: 21	Post Options Post Reply Quote Hammerfest Report Post Thanks(0) Quote Reply Posted: 17 Jul 2020 at 10:57pm
	"I hope there will be updates for new AGESA versions in future and ASRock will provide them to us." - it took a near act of god for getting the current beta BIOS out, I dont have hope THAT SAID, as long as support isn't replying to my tickets, and those of others having issues, with the reply "Its not going to be updated, there is no need to do so" when there is direct guidance from AMD that there is, then I am ok with waiting, delayed VS doing nothing is something rather then nothing. To this point, MSI/Gigabyte have 1.0.0.6 out for nearly all 300 and 400 series boards, while ASUS only has for some highend/midrange boards. Sadly, none of the 6 ASRock board types (several of each) I have in work/client PC's have anything past 1.0.0.4B, it seems ASRock has either a slower, or smaller staff (its normally only 1 guy working on BIOS updates for all systems anyway, or so the reports say) I cant say ASRock has more boards then other vendor's (when you consider the server boards, even Gigabyte has a server board division), but maybe instead of the different BIOS folks like Gigabyte, ASRock has only one for both? Not for sure, I just know that BIOS updates can literally make/break a brand and ASRock corporate doesnt seem to care/know this. If your booting off the Internal and still getting the slow speeds, I will give this a test this weekend, see if I can replicate.

Amiga1000 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 15 Jul 2020 Location: Germany Status: Offline Points: 4	Post Options Post Reply Quote Amiga1000 Report Post Thanks(0) Quote Reply Posted: 19 Jul 2020 at 4:16pm
	100% agree! After all this bad experience with Asrock I'll try a different brand next time. This is quite sad since I like the concept of this little box. With Gigabyte I had problems in the past regarding cooling control. They were famous for these issues but maybe it's history.

m4758406 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 24 Jun 2020 Location: Austria Status: Offline Points: 2	Post Options Post Reply Quote m4758406 Report Post Thanks(0) Quote Reply Posted: 10 Aug 2020 at 7:28am
	Using google translation help cause my english is poor, i am from Austria. My Deskmini nr.3 3400G 2x G.Skill RipJaws SO-DIMM 8GB, DDR4-3200, CL18-18-18-43 (F4-3200C18S-8GRS) ... I can tell you, the RAM (F4-3200C18S-8GRS) was for me incomaptible ! All RAM tests without errors but start problem up to windows. Mostly only the fan ran up, everything else didn't start, sometimes it started up without errors, coincidence. And windows "save energy" sometimes did not work, didn't wake up anymore, frozen black screen. bought new RAM ... G.Skill RipJaws SO-DIMM 16GB, DDR4-3000, CL16-18-18-43 (F4-3000C16S-16GRS) This work great, never had any error since 1 month ! Thank you Asrock.

ASRocker_666 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 06 Jun 2020 Status: Offline Points: 12	Post Options Post Reply Quote ASRocker_666 Report Post Thanks(0) Quote Reply Posted: 20 Nov 2020 at 4:58pm
	These A300 problems are a neverending story: I was very happy to see the new official Bios 3.70 published last week on Nov 11th. With this Bios all my problems regardings graphics are gone (both Linux and Win 10) but I have a déja vu with the boot problems 8 out of 10 times the PC wont come up. The monitor indicates some signal on Display Port but nothing happens. After a few seconds the monitor goes into energy save mode. When I make a hard reset (pressing power button > 3 sec) and turn on again, mostly the PC will start but not everytime. Short term solution for me is to switch back to Beta Bios 3.60K, where the boot problems did not appear. Oh my god ASRock, what the heck are you doing over there? I have NEVER owned such an unmature piece of hardware before! Anybody else having these problems with Bios 3.70 ?

ASRocker_666 Members Profile Send Private Message Find Members Posts Add to Buddy List Newbie Joined: 06 Jun 2020 Status: Offline Points: 12	Post Options Post Reply Quote ASRocker_666 Report Post Thanks(0) Quote Reply Posted: 09 Dec 2020 at 4:46pm
	Just an update on my issue: I downgrade to Beta BIOS 3.60K and the boot failures are gone. Any BIOS above causes problems on my side, the BIOSes 3.60 L to S (AGESA V2 1002) do not work properly on my side as well (graphics failure, and NO there is no RAM defect --> burn in RAM test passed with 0 errors) The BIOS V3.70 has boot problems as described in my previous post. But with 3.60K everything seems smooth. Can someone understand the way ASRock releases BIOS? I do not! First they make an update of the AGESA V1 driver (3.60K), later they introduce AGESA V2 (3.6 L to S) and at the OFFICIAL new BIOS 3.7 they go back to the old AGESA V1 driver but without the modifications they did in the 3.60K version. Really strange politics. ASRock: this sucks !!