EFI Updates
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=4683
Printed Date: 22 Dec 2024 at 2:42pm
Software Version: Web Wiz Forums 12.04 - http://www.webwizforums.com
Topic: EFI Updates
Posted By: mgilbert
Subject: EFI Updates
Date Posted: 25 Mar 2017 at 1:17am
|
" rel="nofollow - I know a lot of you are wondering the same things I am, so I'll ask... I have an X370 Gaming K4 motherboard, running Corsair RAM rated at 3,000 MHz. It was on the QVL, but will only run at 2,400 MHz. Part number - CMK16GX4M2B3000C15. I was able to obtain EFI version 1.70 directly from the ASRock servers, through the EFI. Interestingly, that version is not posted anywhere on ASRock's website. So my first question is... Why isn't the website being kept up to date regarding EFI updates??? In fact, it has been many days since anything new was posted. Most manufacturers are posting regular EFI updates, almost daily. So, my second question is... When can we expect an EFI update that will enable our memory to run at its specified speed? Can a moderator, or anyone, get us answers to these questions???
|
Replies:
Posted By: murakume
Date Posted: 27 Mar 2017 at 4:46am
|
I'd also like to know. I've been refraining from updating through that method since I can't find any documentation on what 1.6 and 1.7 even do. As far as your memory, if they're Hynix memory chips they just aren't playing well right now. I have two sets of TridentZ RGB RAM: 3200C14 and 3200C16. The C14 are Samsung B-dies and the C16 are Hynix; I tried for days on the Hynix to get above 2400 before I gave up and bought the other set. The C14s were plug and play and run at 3200 XMP with manual 1.35 volts. I would hazard a guess that it may be as late as May before we have mass compatibility as that will leave around a month after the rest of the Ryzen product stack releases to see and fix issues. Edit: I'm also on the X370 Gaming Professional, so that may be factoring in also.
|
Posted By: mgilbert
Date Posted: 27 Mar 2017 at 6:26am
| I went ahead with version 1.70, and haven't had any problems. I had hoped this update would help with the memory speed issue. It didn't. In fact, I don't know what changes this update brought, because it STILL hasn't been posted anywhere on ASRock's website. The least ASRock could do is keep their site up to date, and let us know that they are working on updates to resolve the memory speed issues - if indeed they are. |
Posted By: parsec
Date Posted: 27 Mar 2017 at 8:12am
I can answer a question or two, the others are beyond my capability to answer. Regarding your 3000 speed memory, it was learned a few weeks ago from AMD (I wish I had the original source... lost that link, but we' provide it for you) that the Ryzen memory controller (part of the processor) does not support DDR4 3000 and 3200 speed memory. The Ryzen memory controller does not support the "memory dividers" for 3000 and 3200 speed memory. Please do not blame mother board manufactures for AMD not making things like this clear from the start. You may have seen in your board's Memory QVL, that 3000 speed memory will be reduced to 2933. While there is no specification for 3200 speed memory, I believe that will also run at 2933. If the Ryzen memory dividers cannot be changed or subject to an override of some kind, 2933 is the best speed we can get from DDR4 3000 and 3200 memory. http://www.asrock.com/mb/AMD/Fatal1ty%20X370%20Gaming%20K4/index.asp#Memory" rel="nofollow - http://www.asrock.com/mb/AMD/Fatal1ty%20X370%20Gaming%20K4/index.asp#Memory Another factor about Ryzen memory support, is its preference for using a Command Rate (CR) of 1T. With the few existing platforms that support DDR4 memory (Intel's X99, 100 and 200 series chipset/processor boards), a CR of 2T is the universal standard, and using 1T on those platforms is difficult or impossible at higher speeds. This means using the XMP profiles on the existing DDR4 memory will not provide optimal settings for a Ryzen system. Manually setting the memory parameters is the best current practice for Ryzen systems. Regarding why a UEFI version was not posted on a board's web page, but available via other means, I don't know the reason(s) for that. Regarding the rate or frequency of UEFI updates, I don't know anything about that. I took a quick look at the UEFI/BIOS updates for a major mother board manufacture. The GA-AX370-Gaming K7 board has two UEFI versions posted, F2 dated 2/20/17, and F3b dated 3/14/17. The GA-AX370-Gaming 5 has three UEFI versions available, F3 dated 2/13/17, F4 dated 2/20/27, and F5d dated 3/14/17. Do you consider that "almost daily"? ------------- http://valid.x86.fr/48rujh" rel="nofollow"> 
|
mgilbert wrote:Posted By: Trender
Date Posted: 27 Mar 2017 at 8:30am
Well people are getting 3200 MHz with the Corsair CMK16GX4M2B3200C16 and the GA AB 350 Gaming 3 from reading at reddit and YouTube, and the new rymem website http://rymem.vraith.com/detailed_ram/715" rel="nofollow - rymem . And Im about to buy the Asrock X370 Killer or the GA Gaming 3 but Im not sure because of the RAM speeds of ASRock.I dont really care much waiting more as long as ASRock gets its RAM speeds improved too but I dont want to buy a X370 motherboard and being left behind getting stuck at 29XX MHz or lower when I can buy another b350 motherboard for higher RAM speed even with ASRock having more VRMs, RAM speeds are important for me. I still can wait some more before buying so I hope Asrock reach 3200 MHz speeds as well |
Posted By: mgilbert
Date Posted: 27 Mar 2017 at 8:25pm
I'm stuck at 2,400 MHz. As the difference between 2,933 and 3,000 would be imperceptible, I'd be completely satisfied to get to 2,933. Regarding the posted UEFI versions... Again, I was able to get version 1.70 directly through the download feature in the UEFI itself, but that version STILL hasn't been published on ASRock's website, so I have no clue what changes it made. The latest on their website is 1.60, from 3/7. I wish ASRock would keep their website and their server version concurrent, so I wouldn't have to go into the UEFI to check for new versions. From what other people have posted on other forums, I was under the impression that they were getting more frequent updates. Obviously, I was wrong. |
Posted By: wardog
Date Posted: 27 Mar 2017 at 9:55pm
" rel="nofollow -
Do you not recognize the dangers in doing that? From an unknown/unverified source? Then there are your snide, yet unhelpful to yourself, comments of "if indeed they are" and "Obviously, I was wrong." We Mods here are unpaid volunteers. With that comes who I help and how much. Those snide comments of yours do not endear me to your "cause'. |
Posted By: Trender
Date Posted: 27 Mar 2017 at 10:12pm
" rel="nofollow -
Well maybe it wasnt fully tested? Why would they post a update public while theyre working if maybe it isn't 100% tested
|
Posted By: mgilbert
Date Posted: 27 Mar 2017 at 10:57pm
An unknown/unverified source??? I clicked on the "Update UEFI from ASRock's Servers" from within the motherboard's own UEFI!!! How could that possibly link to an unverified source? If ASRock didn't want us using that UEFI update, why would they have a feature in our motherboards that link directly to it??? And when I said, "if indeed they are", I wasn't being snide. I really wonder if ASRock is working on updates that will improve memory speed. Considering that they list 2,400 MHz as the top speed for a dual rank, two piece 16 GB RAM kit right in the specifications for my motherboard, they may not be. And I certainly wasn't being snide when I said, "Obviously, I was wrong". I meant that sincerely. I misunderstood something I had read, and as a result, I made an inaccurate statement, and I was man enough to admit that I was wrong. I appreciate that you are an unpaid volunteer, but perhaps you weren't aware of the feature in the UEFI that links the user to ASRock's updates? And you certainly misunderstood the tone of my other two statements. I would like to be a contributing member of this forum, but if I'm going to be attacked for sincere posts, I'll be glad to delete my account, and never post again. Just let me know.
|
Posted By: mgilbert
Date Posted: 27 Mar 2017 at 11:04pm
Why would they provide a feature right in the UEFI that links you directly to their servers for UEFI updates, if those updates weren't ready. When I clicked that feature, directly in the UEFI, it linked to ASRock's server, and said that update 1.70 was ready. It gave no warning that it might be still in testing, or that it was a Beta version.
|
Posted By: wardog
Date Posted: 27 Mar 2017 at 11:40pm
The underlined above says more of this that I EVER could. Take a deep breath, stop looking for new BIOSes every 5 minutes, and relax. |
Posted By: mgilbert
Date Posted: 27 Mar 2017 at 11:45pm
OK. Won't bother the forums again. Bye...
|
Posted By: wardog
Date Posted: 27 Mar 2017 at 11:57pm
" rel="nofollow -
You're not bothering anything or anyone here in the forum. Do you see where it is I am, and what it is I am attempting to point out? While chasing that "magical' BIOS, what with Ryen being just released and BIOS corrections being issued on a STILL regular basis, when will it be, sooner or later, that you go from complaining of BIOS releases to "Great! The ASRock flash util just bricked my BIOS/board. Who do I bitch at for this happening?" |
Posted By: murakume
Date Posted: 28 Mar 2017 at 7:43am
 On a side note, two new versions of F-Stream have been released in the last week for the Gaming Professional so there is action going on. The download for the non beta version fails as the server appears to be down  but that's beside the point.I agree that, even downloaded through the UEFI, undocumented BIOS updates should be assumed to brick your motherboard. Too much going on at such a basic level to just wish away a failed attempt, you would have to send your board in if you can't witchcraft your way to recovery. It's a new platform, virtually everyone is having issues, it is getting better, and it'll all sort out for the most part by mid April or May.
|