The ASRock thread [support/feedback/etc]

@AsrockMB Can you please let us know what the mosfets/vrm's are on the AB350 Gaming K4 and the PRO 4 please? There is some debate that they are subpar in quality even though you have higher phases than most other manufacturers B350 boards. (And I am keen to get either the K4, or the PRO 4).

Thanks.
Sorry for the delay, but we got a response from HQ (don't have the boards ourselves in Europe yet) - same VRM design on both boards, ISL95712HRZ-T is the controller.
 
Thanks for the feedback on the BX100, @ljt - we've reported that onwards. It does indeed sound like an issue with the AMD controller, but should be solvable with new code incorporated into a BIOS if it works on another controller such as the ASMedia one.
 
No problem, Thanks for forwarding it on. I've also posted it in the thread over on your official forums where there are a few others with the similar issues - http://forum.asrock.com/forum_posts...25463&title=x370-taichi-sata-ssd-goes-missing

It looks as though it's to do with the SSD controller used (in this case the Silicon Motion) rather than one specific SSD.
Had some of our techies look at it, and it's the first they heard of it. Checked with Crucial, who recommend a firmware update: http://www.crucial.com/usa/en/support-ssd?cm_re=us-support-_-main-_-firmware-update-link

Let us know if this helps.
 
Hi. I'm already on the latest firmware for the BX100 (MU02). Do the techs have a BX100 to hand or possibly any SSD with the Silicon Motion SM2246EN controller incase it's to do with the controller itself?

Thanks
We've been in touch with Crucial on the matter - they suggest reflashing to MU02, even if the SSD already has it.

Links:
http://www.crucial.com/usa/en/support-ssd?cm_re=top-nav-_-flyout-support-_-us-support-product-support-ssd
http://www.crucial.com/wcsstore/CrucialSAS/firmware/BX100/MU02/BX100_UPDATE_MU02_BOOTABLE.zip

If you're able to test it for us, we'd appreciate additional feedback.
 
@AsrockMB can you suggest any rgb light strips that work well with the taichi board ?
From the top of our head: Bitfenix Alchemy 2.0 RGB is a nifty product. We haven't tested too many (as in us handling these forums), but we have fiddled with those - the magnetic mounting make them very easy to deal with, and they're available in different lengths as well. We're sure other users have other suggestions (and more experience). :D
 
i have a small complaint

the taichi deserves more gold badges :p
We welcome you to place additional stars on the table as you see fit. :D

Just one complaint, the VRM phase count is an issue here. The 9 phases sounds great until you find out that it's a 3 phase on the SOC and a "6" phase on the Vcore. It's not actually a 6 phase, it's a 3 phase with too many components. Ie: two inductors, two low and two high FETs per phase. The reason being is that the ISL95712 voltage controller has two Vcore phase drivers and 1 northbridge phase driver (used for SOC here) integrated into it. The Vcore section of the board only has 1 phase driver on it. Therefore it is not a 6 phase, it's a 3 phase. Due to the voltage controller having integrated phase drivers it is impossible to using a doubling scheme and there are no doublers on the board.
This means advertising as 9 phases is strictly incorrect, there are exactly 6 phases on the board, 3 for SOC and 3 for Vcore. It just looks like a 6 phase due to having extra components.

All this is taken directly from this video and have looked at the data sheet for the ISL97512 and have confirmed that this is the case. A 6 phase Vcore with it is not possible.

Now I understand that it will power these CPUs just fine especially the R5 and below CPUs. There is no real problem with the power delivery just the advertising which seems intentionally misleading and intended for pictures like these^^ to make it look better. Advertise it in a way that makes dual FETs sound really good but don't claim the board has 9 phases.
It's not unlikely that this is a mistake where someone's equalled chokes to phases. Like you say, it's a 3-phase system with dual chokes, and we hadn't given it much thought before you brought it up. We'll forward this to HQ.

It is completely fine for all current CPUs though, and these lower-end boards with this VRM design aren't designed with overclocking in mind.


@AsrockMB
hello again
we seem to have an issue with the samsung 960 nvme drives and the 4K performance being very low (there is a thread on the asrock support forum about this ((unresolved)) ) i wonder if you can ask the question to engineering ?
it seems very slow on asrock boards compared to 'other' brands at present its two taichi owners (same evo 960, same board and bios) and one k4 x370 owner with a evo 960

here is a screen shot showing my evo 960 on nvme main slot, barely beating my evo 850 on my sata connections at 4k


thank you in advance
We'll forward this. Currently there's a long weekend (May 1st is off in Taiwan), and yours truly is also moving on from ASRock, but our tech guys from The Netherlands will most likely handle these issues in the short term. Apologies in advance if there's an additional delay of a couple of days in getting back to you, or anyone else in the thread.

@AsrockMB - having a look at the Memory QVL list and I'm not sure if my memory is deceiving me, but have most of the G.Skill Trident Z kits been removed from the list? I swear I saw quite a few on there but now I can only see two. There was definitely a 3200MHz Trident Z RGB kit on that list, I'm sure of it.
For which board? It is possible some kits have been removed as memory testing and optimization is an ongoing process for the Ryzen platform.
 
Back
Top Bottom