Biostar Racing X370-GTN ITX AM4

Does X37AK919.BSS feature any feature changes that anyone has noticed? Are overclocks behaving different in terms of voltage down-clock or anything. Has p-state overclocking returned? Or is this just an agesa stability revision?
 
richiegore said:
Does X37AK919.BSS feature any feature changes that anyone has noticed? Are overclocks behaving different in terms of voltage down-clock or anything. Has p-state overclocking returned? Or is this just an agesa stability revision?
Click to expand...
No, unfortunately p-state overclocking has not returned, and we're still stuck with the voltage offset not going below 1.35V, which is way too much. It appears be just the 1.0.0.6b revision, which only fixes a rare Ryzen bug as far as I know.


richiegore said:
Supposedly there is connection between your CPU and what memory speed you can hit. Maybe you could try up the SOC voltage? I think that is what controls infinity fabric which connects to the ram... I haven't done any research into it but it's just something I have seen mentioned. Might be worth looking into.
Click to expand...
Does anything actually happen when you up the SOC voltage? I did try that in the past, but I remember that any voltage monitoring utilities, including Ryzen master, kept reporting the same voltage regardless of the BIOS setting. Maybe I should look into this again, because I could be wrong. Thanks for your advice.
In any case, the whole voltage offset way of doing things is cumbersome. I wish you could just dial in a specific voltage.
 
klm said:
No, unfortunately p-state overclocking has not returned, and we're still stuck with the voltage offset not going below 1.35V, which is way too much. It appears be just the 1.0.0.6b revision, which only fixes a rare Ryzen bug as far as I know.



Does anything actually happen when you up the SOC voltage? I did try that in the past, but I remember that any voltage monitoring utilities, including Ryzen master, kept reporting the same voltage regardless of the BIOS setting. Maybe I should look into this again, because I could be wrong. Thanks for your advice.
In any case, the whole voltage offset way of doing things is cumbersome. I wish you could just dial in a specific voltage.
Click to expand...
Wow, 1.35 that is a high minimum. I think I'm going to try Ryzen master, see if I can get one core up to 4.0 or 4.1 and leave the rest as is... I reckon that will give the best gaming performance while keeping temps under control. Not sure how good Ryzen master is at down-clocking and down-volting though
 
I use Ryzen Master to set my overclock. It's a bit of a PITA that you need to redo it after every boot-up, but to keep noise down, I don't want to run at 1.35V. I run at 3.8 GHz / 1.25 V. If, in the future, I do set my overclock in the BIOS, I might as well aim a little higher, if the voltage is at 1.35 V anyway. When we still had p-state overclocking with the first BIOS, the minimum was 1.275 V which was ok.
 
klm said:
I use Ryzen Master to set my overclock. It's a bit of a PITA that you need to redo it after every boot-up, but to keep noise down, I don't want to run at 1.35V. I run at 3.8 GHz / 1.25 V. If, in the future, I do set my overclock in the BIOS, I might as well aim a little higher, if the voltage is at 1.35 V anyway. When we still had p-state overclocking with the first BIOS, the minimum was 1.275 V which was ok.
Click to expand...
That's a good overclock at that voltage. I've been looking at the digital foundry results here and a 1700 with 3200 ram is faster than an 1800x with 2133 MHz ram. Of course an 1800x with 3200 ram is faster again but still... As I have managed to get 3200 on my ram I think I'm happy enough to keep the 1700 stock unless I start doing some serious 3D rendering again.

On a side note I couldn't find a way to clock one core faster than the rest in Ryzen master. Am I missing something or is that just how it is?
 
Minimum voltage depends on procesor model, for my 1500x I can set it to 1.235 lowest using lowesr offset setting.
 
Been playing with bclk overclocking, set it to 104.3 and it does boost to 3.9 but only for a second then it drops back to 3.3... keeps jumping up to 3.9 every so often. Temps seem fine, any idea why it's not staying at 3.9? I think bclk could be a good way to get a high boost like a 1800x... I got 147 in cinebench single threaded.
 
Ok so I'm having trouble with my SSD not being detected on every boot. I thought previously it was an issue with the crucial mx100 drive I had, so I swapped it for a Samsung 840 evo. The evo had been great for ages but the problem seems to have returned.

I think maybe the sata cable is damaged but before I try to swap it I thought I'd ask to see if anyone is having a similar problem?
 
I had a similar problem with my build - though since I was using an M.2 for storage, the sata that was disconnecting was the optical / BR drive. The m.2 never had a problem the disappearing optical drive was annoying. The second issue for me was that the mobo audio would be staticky the entire time / no matter what I did, so I had to turn it off and run a USB soundblaster solution instead. Third, when I started having problems with the mobo not being able to keep the XMP (it would start acting wonky unless I ran it at stock clocks), I finally pulled the trigger on the current mobo - which works flawlessly.
 
dvsman said:
I had a similar problem with my build - though since I was using an M.2 for storage, the sata that was disconnecting was the optical / BR drive. The m.2 never had a problem the disappearing optical drive was annoying. The second issue for me was that the mobo audio would be staticky the entire time / no matter what I did, so I had to turn it off and run a USB soundblaster solution instead. Third, when I started having problems with the mobo not being able to keep the XMP (it would start acting wonky unless I ran it at stock clocks), I finally pulled the trigger on the current mobo - which works flawlessly.
Click to expand...
Hmmm... That's interesting, i still hope it's the cable, I may have bent it too hard routing it in my Dan case.... what motherboard are you using now btw?
 
dvsman said:
It's the second rig in my sig - Asrock x370 itx/ac
Click to expand...
Cool thanks, I found it on your profile page, I think the profile page could be improved a bit on hardforum... I swapped the cable and it's been ok since, been rebooting and switching it on and off all evening. It's far to early to say yet though... The cable was routed behind the graphics card so that may have been the problem, probably a load of interference from the GPUs VRMs maybe... I have routed it differently now.... I will post if the problem returns!
 
richiegore said:
Ok so I'm having trouble with my SSD not being detected on every boot. I thought previously it was an issue with the crucial mx100 drive I had, so I swapped it for a Samsung 840 evo. The evo had been great for ages but the problem seems to have returned.

I think maybe the sata cable is damaged but before I try to swap it I thought I'd ask to see if anyone is having a similar problem?
Click to expand...

No issues since I changed the cable but I think the issue was actually that I had the cable routed behind the graphics card in my DAN-A4. I also changed the cable route when I changed the cable. I think there was probably a bunch of interference from the GPUs VRMs. Pretty silly in hindsight!
 
Did anybody try the new BIOS yet with AGESA 1.0.7.2a?
 
klm said:
Did anybody try the new BIOS yet with AGESA 1.0.7.2a?
Click to expand...

I did update this, but It didn't make too much change for me - RipJaws V 3200 are still not kicking in on full XMP and I have to run them slightly clocked down to 2933, Boot up to POST time on overclock is still long, and failed overclocks are still making you wait 5 boots until they reset the config completely.

One interesting thing to notice is that I didn't reset the overclock before updating the bios and maybe because of that my CPU voltage was improperly configured on the first run showing the ramped up value without + voltage setting being configured and thus after saving OC back it failed and had to be reset. So you might want to set no overclock on the first save or clear your setup before flashing.

Should I check a specific feature?
 
BIOS in this MoBo is just a nightmare.
Have they added undervolting option?
 
SaperPL said:
Should I check a specific feature?
Click to expand...

I read somewhere online that this new AGESA has made it possible on some motherboards to disable AMD's PSP feature (comparable to Intel's ME). I will update later and search for it in the settings, but to be honest I doubt Biostar does any real effort to improve this BIOS.

Thanks for the tip on resetting the overclocking settings before updating.
 
I think after this update my system is slightly unstable - I get some chrome crashes from time to time and I get to wait for webpage to load while freezing chrome from time to time as well. I'm not recommending this revision. Wait for b revision, hopefully they'll catch up quickly to that.
 
I just did the update. RAM @ 3200 MHz still won't boot sometimes, for the rest no change.
 
SaperPL said:
I think after this update my system is slightly unstable - I get some chrome crashes from time to time and I get to wait for webpage to load while freezing chrome from time to time as well. I'm not recommending this revision. Wait for b revision, hopefully they'll catch up quickly to that.
Click to expand...
I've had this board since March '17, got in early.. it's ok, I actually like it. Stable for me. Fan controls are horrendous though. Basically had to buy the fan that works best rather than tune the one I have. I did have to rollback from that bios (AGESA 1.0.7.2a update). X37AK919.BSS (AGESA 1.0.0.6b update) is the one I use and works well. Supposedly X37AKB30.BSS contains the memory rewrite and I had to blind flash back to X37AK919. No video until it got into Windows no matter what output I had plugged in. Wouldn't boot at all with any monitors plugged in at all. Made me miss analog display outs.
I've been running into all sorts of bad updates lately. Latest NV driver that works well for me is 388.71.

Not touching this board's bios for at least 6 months with all the disarray over Meltdown/Spectre. Probably move to an X470 or X570 board though at a point.
 
housecat said:
I've had this board since March '17, got in early.. it's ok, I actually like it. Stable for me. Fan controls are horrendous though. Basically had to buy the fan that works best rather than tune the one I have. I did have to rollback from that bios (AGESA 1.0.7.2a update). X37AK919.BSS (AGESA 1.0.0.6b update) is the one I use and works well. Supposedly X37AKB30.BSS contains the memory rewrite and I had to blind flash back to X37AK919. No video until it got into Windows no matter what output I had plugged in. Wouldn't boot at all with any monitors plugged in at all. Made me miss analog display outs.
I've been running into all sorts of bad updates lately. Latest NV driver that works well for me is 388.71.

Not touching this board's bios for at least 6 months with all the disarray over Meltdown/Spectre. Probably move to an X470 or X570 board though at a point.
Click to expand...

Yeah, I'm also planning to ditch this board for X470/B450 from Asrock or Gigabyte, possibly along my current memory sticks and maybe use that leftovers for Ryzen APU tests when it launches.
 
  • Like
Reactions: RosaJ
like this
I ditched mine when other mobo options came out. I bought in when it first came out as well but my early version apparently had scratchy / staticky audio that just couldn't be fixed. Then it developed problems on the USB ports ... so the switcheroo to Asrock board - which has been rock solid and now everything works as advertised.
 
Agreed, I actually held out because of this forum and went with the ASRock AB350. 10/10, love the board. The bios and fantastic ram OCs is my main takeaway. Was able to get my Team T-Force ram down to 3200 w/ 14-16-16-32 timings.

Post from earlier:
RosaJ said:
Currently have Team T-Force XTREEM 16GB (2 x 8GB) DDR4 3733 in my Sentry case for 169.99 USD now 196.99 USD, jeez those prices! Currently have them at 3200 MHz with tight timings as I didn't want to go higher than 3200 MHz with loose timings.

CPU-Z:
N52QHQ9.png


Edit: My mobo is the ASRock AB350itx and CPU is R7 1700X
Click to expand...

So thank you for your sacrifice and posting your troubles!
 
I'm still running my board without issue. CPU stock, ram at 3200. I don't think I'll bother changing it.
 
I didn't mean to imply that I was replacing my board either. I like the board, it's rock solid stable. People playing outside the official memory specs are asking for problems. I use XMP or bone stock anymore. The only guarantee is 2667MHz/2400MHz/2133Mhz tops, those are the Ryzen specs. I'm using the XMP profile on my 32GB DDR4-3200 kit, except bumping the memory clocks down to the official spec for my config. I may move to X470 only because if I'm changing the CPU/heatsink or any part I may as well do more.

I will say the board is definitely a throwback to old school days, and I would only recommend it for those who consider themselves well-grizzled experts who feel they know what they're doing. It's definitely not going to hold your hand like the more polished stuff from Asus. On quality, polish and support the market is really divided in two tiers, Asus, then everything else. If I'm buying today, I'm going with that just because it's what I've used for 20 years.

Ultimately, times have changed and everything works unless you really got something else screwed up as CPUs have trended towards SoCs.
 
No thanks on that beta bios when the last one was unstable. I’ll give them a year before updating again.
 
Updated my board today at the latest bios version. PC boots with 2 sets of three fast beeps and black screen.Be warned, the update might bork the board. I am trying to figure out how to solve it, haven't figure this out yet, I will keep you posted.
 
PlanDreaM said:
Updated my board today at the latest bios version. PC boots with 2 sets of three fast beeps and black screen.Be warned, the update might bork the board. I am trying to figure out how to solve it, haven't figure this out yet, I will keep you posted.
Click to expand...
Hey dude I feel your pain, let me try to help you. I had issues on any UEFI newer than the AGESA 1.0.0.6b update from 2017-09-19, especially with video card handshaking / initialization, but I got it figured out.
  1. Unplug everything from the computer, including the monitor(s). Turn the machine on, after it would normally be at the desktop, plugin a monitor to see if it detects it. If so, that means you're definitely in a salvageable spot.
  2. While you're there, put either X37AK919.BSS or X37AK309.BSS from http://www.biostar.com.tw/app/en/mb/introduction.php?S_ID=879#download on a USB drive. Your pick. X37AK919.BSS is the safe, guaranteed to work choice. I use X37AK309.BSS since it's the latest UEFI (& supports Ryzen 2000 APUs). We're going to do a blind flash.
  3. I recorded this video for myself before going to the latest UEFI, now it's on Youtube. This will help you with the keys to press, and general time it takes to complete so you're not left sitting there, becoming concerned.
  4. Once you get it back, if you chose X37AK309.BSS you'll probably want to leave the video sections of the BIOS CSM options as per the screenshot attached to this post. With X37AK919.BSS you shouldn't have to worry about anything, I never had any issues with that release.

I actually find this board is decent, it's not for novices though. This is also not the only motherboard manufacturer struggling with wonky updates. Most of this is coming from AMD, though the validation failures can be pinned on the board vendors for sure. Be interested to hear where you end up! I hope the steps/video/screenshot help you. Good luck.
 

Attachments

  • IMG_0270.JPG
    IMG_0270.JPG
    310.7 KB · Views: 0
housecat said:
Hey dude I feel your pain, let me try to help you. I had issues on any UEFI newer than the AGESA 1.0.0.6b update from 2017-09-19, especially with video card handshaking / initialization, but I got it figured out.
  1. Unplug everything from the computer, including the monitor(s). Turn the machine on, after it would normally be at the desktop, plugin a monitor to see if it detects it. If so, that means you're definitely in a salvageable spot.
  2. While you're there, put either X37AK919.BSS or X37AK309.BSS from http://www.biostar.com.tw/app/en/mb/introduction.php?S_ID=879#download on a USB drive. Your pick. X37AK919.BSS is the safe, guaranteed to work choice. I use X37AK309.BSS since it's the latest UEFI (& supports Ryzen 2000 APUs). We're going to do a blind flash.
  3. I recorded this video for myself before going to the latest UEFI, now it's on Youtube. This will help you with the keys to press, and general time it takes to complete so you're not left sitting there, becoming concerned.
  4. Once you get it back, if you chose X37AK309.BSS you'll probably want to leave the video sections of the BIOS CSM options as per the screenshot attached to this post. With X37AK919.BSS you shouldn't have to worry about anything, I never had any issues with that release.

I actually find this board is decent, it's not for novices though. This is also not the only motherboard manufacturer struggling with wonky updates. Most of this is coming from AMD, though the validation failures can be pinned on the board vendors for sure. Be interested to hear where you end up! I hope the steps/video/screenshot help you. Good luck.
Click to expand...

I am trying it right now! Thank you. So the 6 beeps are the VGA initialization? Did you have the same problem with the beeps?
 
PlanDreaM said:
I am trying it right now! Thank you. So the 6 beeps are the VGA initialization? Did you have the same problem with the beeps?
Click to expand...

My machine doesn't produce any audio codes, ever. I have them enabled, but unsure if due to speakers plugged into the wrong port or if it's my case not having a traditional PC speaker built-in. It's an AMI UEFI so if you look for that online you'll find what the different codes mean, should also be in the manual. I looked but seems odd if it's really a keyboard failure, unplugging everything will reveal if there's a solution.
Beeps are good though! Something to work off of.

If it is video like what I ran into, you may need the UEFI video initialization instead of legacy BIOS as my system requires. That means you might have no choice to use X37AK919.BSS, since these all default to legacy upon a clean flash. Unless you go into the UEFI blind and change the CSM settings (a bit too much work for me, I'd just use X37AK919.BSS).
Also, I always reset to defaults/optimized defaults before a flash, helps prevent some strange issues from happening.

edit to add- took a quick look at your profile since you're a new poster. Jun 10, 1982 (Age: 35). You're two days older than I am. From one (relatively, for PC nerds), old balls to another, I hope you get this squared away.
 
Yes, it does beep and when I take the memory out it beeps correctly: 3 times fast. I reset all settings as well before a flash, seems to help.

In order to hear the beep I installed a beeper on the header

Edit: doesn't change beeping when I take out the VGA, which seems to be a good thing from what you said.
 
Last edited:
The 6 beeps reads to be possibly a little dire. https://www.lifewire.com/amibios-beep-codes-2624543 If you can't get it to flash back to X37AK919.BSS (which is what I'd go for if I were in that situation because it's solid and doesn't introduce new problems), blind or otherwise, it's probably dead. In which case you could at least RMA or take the opportunity to move to the Gigabyte or Asus boards.
Given the 6 beep error, be sure to test without the keyboard/mouse connected to see what happens.
 
Last edited:
Yeah, I have seen that too but it might not be the GateA20 but something else. I am waiting for BIOSTAR and the seller to reply.
 
Did they upload the new revision or did they just remove the "Beta" from the name?
 
I am running the latest X37AK417.BSS BIOS (AGESA Pinnacle PI 1.0.0.2a update), after I read elsewhere that people found out it improved memory stability for their first-gen Ryzen systems on various mobos. It appears I can finally get my RAM stable at 3200 MHz at rated timings, albeit at 1.40 V. SoC voltage is still at AUTO. I have run various benchmarks and memtests, all passed, and so far it has not once failed to boot and I have done a lot of restarts. Just loading the XMP profile also worked fine during normal usage, but I got an error sometimes during memtesting, so I upped the voltage to 1.40 V and now the errors are gone at the advertised timings, which are slightly tighter than what I get if I just load the XMP profile.

Now I sincerely hope it will stay this way!
 
I'm also using X37AK417.BSS BIOS (AGESA Pinnacle PI 1.0.0.2a update) now. No problems here that didn't exist before, but I don't overclock. Every release after X37AK919.BSS AGESA 1.0.0.6b update doesn't show video output on my GPU unless I put it in BIOS mode in CSM.
Only issue I have with this system as it is, after sleep or hibernate I get Reset to device, \Device\RaidPort1, was issued. I've done everything conceivable over months to resolve it, different NVME drivers, tested most UEFI releases, replaced with a new drive, power options, you name it. May test again on Win10 1803, because I don't recall this issue happening on 1703.

So with S3 sleep disabled in UEFI, no problems at all. Other than that, even with BIOSTAR & AMD at the wheel (what a combo right), I've had a pretty good experience. Will probably replace my CPU/motherboard with 7nm Ryzen.
 
You must log in or register to reply here.
Back
Top