LLC on the Zenith Extreme (Part 2)

Message boards : Number crunching : LLC on the Zenith Extreme (Part 2)
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1952332 - Posted: 27 Aug 2018, 10:29:02 UTC

I originally posted my observations of LLC behavior of the Zenith Extreme with a 1950x in this thread. I found that none of the LLC Levels where anywhere near flat and that when I optimized Vcore for full load, the system would crash under light load. To avoid this, I had to increase Vcore by 25mV.

In preparation for upgrading to the 2990WX, I had updated the BIOS to 1402 over the weekend and noticed 2 new LLC Levels. I repeated my previous DOE by measuring the impact of LLC Level on measured Vcore (ProbeIT terminal) and found the 2 new levels addressed my originally concerns. The new -L1 setting resulted in a nearly flat curve with 100% loading only 4mV higher than low loading. The -L2 setting results in some droop with 100% loading 11mV lower than low loading.


I am now running with -L1 at original set voltage, which results in Vcore being 25mV lower than L1 at full load and nearly identical Vcore at low loading. It has ran for 26hours with no issues. I will report back on how stable it is during the outage and if it maintains long term stability. I plan to upgrade to 2990WX which I will purchase during my US trip planned in Sept.
GitHub: Ricks-Lab
Instagram: ricks_labs
ID: 1952332 · Report as offensive
Ian&Steve C.
Avatar

Send message
Joined: 28 Sep 99
Posts: 4267
Credit: 1,282,604,591
RAC: 6,640
United States
Message 1952364 - Posted: 27 Aug 2018, 17:00:51 UTC

mind, or address, your VRM temps if at all possible. that will probably be a limiting factor with so much power pumping into a 32core CPU. direct airflow would be ideal rather than the passive sinks on there.

if you could get a monoblock and watercooling, that would be best i think.
Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

ID: 1952364 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1952388 - Posted: 27 Aug 2018, 18:59:16 UTC

I don't remember anyone mentioning new or improved LLC in the 1402 BIOS in the OCN ZE thread. Most of the comments are about ASUS finally fixing the fan control and the still ongoing issues with RAID.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1952388 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1952450 - Posted: 28 Aug 2018, 0:07:44 UTC - in response to Message 1952364.  

mind, or address, your VRM temps if at all possible. that will probably be a limiting factor with so much power pumping into a 32core CPU. direct airflow would be ideal rather than the passive sinks on there.

if you could get a monoblock and watercooling, that would be best i think.


Yes, I have read that this MB support of the power needed by 2990WX is marginal for OC'ing. I am running with the EK mono-block and don't plan to OC, so I should be ok.
ID: 1952450 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1952452 - Posted: 28 Aug 2018, 0:15:33 UTC - in response to Message 1952388.  

I don't remember anyone mentioning new or improved LLC in the 1402 BIOS in the OCN ZE thread. Most of the comments are about ASUS finally fixing the fan control and the still ongoing issues with RAID.


I imagine that most people don't share the concern I raised about the previous LLC options. Seems like people are happy to set a safe OC Vcore and assume it results in a safe Vcore.
ID: 1952452 · Report as offensive
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1952462 - Posted: 28 Aug 2018, 0:56:51 UTC - in response to Message 1952452.  

Not that we don't care, its just we can only dream of a TheardRipper ...
ID: 1952462 · Report as offensive
Profile RueiKe Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 14 Feb 16
Posts: 492
Credit: 378,512,430
RAC: 785
Taiwan
Message 1952475 - Posted: 28 Aug 2018, 2:31:31 UTC - in response to Message 1952462.  

Not that we don't care, its just we can only dream of a TheardRipper ...

When I wrote “most people”, I definitely didn’t mean this community. We are a bit unique in the community like an Asus forum. Some issues found in our use model would impact very few people outside of this community and often don’t get a lot of attention. Not complaining, as we do benefit substantially from innovations driven by the gaming community, but we are on our own for some unique issues.
ID: 1952475 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1952482 - Posted: 28 Aug 2018, 4:24:38 UTC

I still think I would like to build the TR system at some time just to get familiar with that many cores. I think my experience with Ryzen gives me a fairly solid understanding of the architecture. I have been reading the ZE OCN thread since it started and lurking trying to absorb the knowledge divulged there for possible future use someday. I don't see myself with a 32 core monster but think I could use the the 16 core chip. But I would want the new 12+nm feature size since I've seen first hand the benefit of the 2nd generation over the first.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1952482 · Report as offensive

Message boards : Number crunching : LLC on the Zenith Extreme (Part 2)


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.