AMD X399 / Threadripper motherboards appear

Asus usually launches their WS stuff a bit later than all of their other boards. We are still waiting to see a X299 WS series from them as well, so I would guess it will be a bit.

1 Like

Hi Drew - well I have an X99-E WS/USB3.1 working flawlessly in FreeBSD (FreeNAS); didn’t try linux on it though; which board do you have? Is it the WS 10G board?

What sort of issues are you facing?

Actually, I ain’t facing any now, but my system did have some trouble reading the USB ports.

One problem I do have with the mobo on both Windows and Linux is the sound. It’s not stereo when it should be.

nice board…similar to X370 …must be a "Taichi"design.

I’ll wait for X399 Sabertooth from ASUS. And MAYBE I’ll finally upgrade my X79 to X399 (1900X).

Same here, rocking my X79 sabertooth with a E5-2670 in it.

Then a coin flip for the 1900x or 1950x when I do upgrade.

Msi X399 Gaming pro carbon vrm spec’s.

Main Vcore vrm:

  • pwm: IR35201 providing 5 phases that are being doubled to 10 using IR3599 phase doublers.
  • powerstages: IR3555 which are rated for 60A at 125°C
  • inductors: unknown but my guess their rating are arroung 60A / 65A ish.
  • caps: Tantalum’s.

Soc vrm:

  • pwm: IR35204 providing 3 phases for the soc vrm.
  • powerstages: IR3555´s again.

Memory vrm:

  • pwm: 2x Primarion PV4210 (IR) digital pwm, providing 2 phases each for each set of 4 dimm slots.
  • powerstages: Texas Intstruments CSD87350Q Nexfets rated upto 40A at 125°C.

Well done Msi this is a pretty decent vrm.

1 Like

Asus X399 ROG Zenith Extreme vrm:

Main vcore vrm:

  • pwm: ASP1405, which is “most likelly” a rebranded IR35201,
    providing 8 true phases for the main vcore vrm.
  • powerstages: IR3555M which are rated upto 60A.
  • Chokes / inductors: Microfine alloy chokes 60A.
  • caps: Tantalum poscaps and 10K caps.

Soc vrm:

  • pwm: ASP1405 providing 3 phases for the soc vrm.
  • powerstages: Texas Instruments CSD97374Q4M which are rated upto 25A.

Memory vrm:

  • pwm: 2x ASP1103 providing 2 phases each for each set of 4 dimm slots.
  • mosfets: Each phase gets a single High-side and single low-side fet.
    but i dont know which those are.
1 Like

How can I get a rundown on the Asus Prime X370 Pro? :wink:

You mean the vrm on that board?

Yes, mostly out of curiosity since I own one and am currently running it OC’d, but it never seemed to get much attention compared to most of the high end boards.

I have not looked at that particular board myself yet.
But what i could find about it, and based on pictures.

  • pwm: ASP1405 / IR35201 running in 6+2 phase mode.
    6 straight phases are used for the main vcore vrm.
    And the +2 phases are being doubled to 4 for the Soc vrm.

  • powerstages: Ti Nexfet CSD87350 40A.

  • inductor ratings: unknown.

  • caps: 10K.

This is what i´m assuming it to be.

1 Like

Interesting video regarding X399 in its current state (title notwithstanding):

So far this is the only video (of YT channels I’m subscribed to anyway) regarding issues with the platform, everything else has been pretty quiet.

As someone who has worked in tool & die making progressive dies I’m really surprised to see the variation in the socket manufacturing, and I would hope AMD catches wind of this and comes down on them hard enough to make a difference. This shouldn’t be happening at all.

Much of the rest seems to be motherboard manufacturer issues, software hangups, new platform type stuff.

Any early adopters want to comment? @wendell perhaps?

1 Like

Haven’t watched the video yet but it does have teething issues. Aside from iommu/virtualization it has been rock stable for me on Linux. I am waiting on gigabyte to resolve a couple issues before I publish my review. No word from Asus on anything Ryzen or threadripper, and ASrock has had some similar issues but updated.

The MSI folks have been very communicative on issues , status, and feedback from me on issues. I get the sense they are passing it back up the chain with amd.

Had trouble with 128gb kits of memory that is mostly fixed now on the MSI carbon pro ac.

2 Likes

The socket is tricky and maybe why he has issues.

First install I had zero issues. 3200MHz 64GB in 4x 16 sticks memory posted at 3066 with DOCP but not 3200. That was with zero tweaking. Ran Windows install and zero issues. CPU was seated easily and did not have any issues he speaks of. So all in all very satisfied.

Then I changed the CPU.

Would not post quad channel, stability issues and getting to Windows was at times problematic.

Removed CPU and this:

Nothing came in contact with the socket other than the CPU. Sent MB to Asus for repair.

Advice: Seat CPU on a flat surface outside the case and follow these instructions (note he mentions pins “moved”):

So same motherboard, two different CPUs, two totally different outcomes.

Tech Showdown? Didn´t that guy stir up some shit half a year ago that then turned out to be a non-issue?
The points made are Foxconn problems (or Gigabyte and Asus), not AMD problems. Nonetheless they need to be resolved.

Moral of the story: Read reviews first. Gigabyte has been a bit hit or miss since 1366 socket days.

Oh I agree completely, hence my insight via manufacturing. I have made progressive dies that make parts like the hinged lids on the sockets, and it appears to me as though Foxconn has bad tooling. This is common in Chinese manufactured tooling by the way, and also why there is still a lot of progressive die manufacturing in the US that is then sent overseas for production purposes (cheap labor with good tooling).

1 Like

Yes, his title is click bait for sure. A wannabe reviewer. His buddy Tech YES City is an arrogant smock too…

Be interesting to see if ASUS repair/replace that board.

Interesting, but of course Threadripper is a new platform,
so its not that strange that there some minor bugs.
The socket issues he talks about is interesting though, thats definitelly something to do some more research into.