Unsolved
This post is more than 5 years old
5 Practitioner
•
274.2K Posts
0
1443
Alienware X51 R3 repeatable Power Reboot during certain condition (GTX960)
I having this repeatable issue where the X51 R3 + GTX 960 reboot without kernel dump when under workload. (Windows 10 + Final Fantasy 14 application (benchmark)
After contacting Support and replacing almost every component and still observe the repeatable issue. Support was unable to perform RCA.
When prompt if there is any known issue with X51 R3 having under power rating of 330 watt when GPU request for system to be at least 400 Watt requirement. Support denied at every point.
Having experience this issue for 6 months and no other solution, i resort to purchase External APU at my own expense and issue was resolved.
For customer who has X51 R3 and having powerful GPU, Please consider APU add-on. I would have to (sadly) deduce that the validation for the GTX960 for this platform was not done accordingly.
OS Name Microsoft Windows 10 Pro
Version 10.0.14393 Build 14393
OS Manufacturer Microsoft Corporation
System Manufacturer Alienware
System Model Alienware X51 R3
Processor Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz, 3401 Mhz, 4 Core(s), 8 Logical Processor(s)
BIOS Version/Date Alienware 1.2.8, 8/30/2016
SMBIOS Version 2.8
Embedded Controller Version 15.07
BIOS Mode UEFI
BaseBoard Manufacturer Alienware
Hardware Abstraction Layer Version = "10.0.14393.206"
Installed Physical Memory (RAM) 16.0 GB
NVIDIA GeForce GTX 960
89fordprobee
1.2K Posts
0
February 8th, 2017 07:00
okay recommendations for power are based on average use add on accessories and keeping load under 80 % of the psu kinda thing..
now that said I have a 6700K and a 1070 32 gig ram and such on my X51 R3 and I much run it at 90% plus 7-12 hours a day and have no issue ... it draws REAL WORLD off the wall with a 32 inch 2k screen 270~ watts according to backup power screen .. which is roughly 65- 70% of psu .. after taking screen in account power draw
dell also sold the X51 with a 6700 and a 970.. OEM cards aren't the same as retail they have set specifics for power restrictions aswell so your conclusion isn't accurate :(
if parts are replaced and under warranty request replacement system, if everything has been changed and still continues its possible the refurb parts your getting have an issue.. they don't go through stress test when tested.. just that they turn on and function.
I know when I got my R3 it had a HDD DOA ... and also had overheat issues with the stock 6700 (NON K model cpu) ended up after disassemble of it hitting boiling point of water I found out that no thermal pad/paste was on cpu .. kinda disgrace but it happens. the original 6700 and 970 are sitting in a tower beside me now ..
as for your issue with external video fixeing it, leads me to believe its video card/daughter board issue, power board issue or WRONG power supply/defective or just straight out overheating ....
Tesla1856
8 Wizard
8 Wizard
•
17K Posts
0
February 8th, 2017 11:00
I have a couple APC Back UPS XS 1500 (LCD) with similar wattage display. Also similar APC-1000-LCD units.
I don't think it's wattage-readout is all that accurate. Seems too low to me.
Either that, or the 120v wattage is not the same as the 19v wattage the X51's "laptop-like" power supply must output. Maybe a lot of power is lost in this DC-DC Converter arrangement. There is also the tendency for power-supplies stability to drop-off as it nears 100% utilization.
I've often thought the x51's power-supply was undersized for the nicer cards. I think the problem comes at peak-usage (max gaming). Users have done various experiments proving this, including hooking up a nice power-supply outside the x51 and/or moving everything to a new case with conventional desktop power-supply inside.
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
February 9th, 2017 18:00
I have to agreed to some of the points however I doubt some of the conclusion is affecting me as my observed behavior is where system cut power to the board and reboot afterward. No dump no log.
All Therm data show it way below 85C well within specification.
I also try to use 3rd party application to cut GPU power limit and CPU power limit to 90% however then it come to my stress test (FF14 Benchmark) at stage 2 it always exhibit the behavior.
As you mentioned the result is inconsistent and hard to point the fault.
oh i also forgot to mention i had the system driving the u3415w monitor at native resolution and 2560 x 1080 resolution on the app.
Anyway as far as my case the external APU currently has been running for past 10 hour loop on the test app. it has not exhibit same issue since. finger cross.
Tesla1856
8 Wizard
8 Wizard
•
17K Posts
0
February 9th, 2017 18:00
1. If it were a software problem, is usually writes to log. However, with a general hardware problem or running out of power/amps/watts ... it usually just turns off (no logging).
2. I'm guessing you mean external Advanced Graphics Adapter (AGA). Yes, this makes sense as you are introducing more power to help run the system as a whole. In this case, AGA powers the 960, leaving the 330 watter to power the X51 itself.
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
February 9th, 2017 19:00
You are right on point 1. I had to wipe my entire OS because the Support refuse to acknowledge the issue was not software.
Good new i can boost beyond 960 now.
Reason i post this thread was i could not find any official communication or acknowledgement from the Support on the under power issue. hence hopefully if anyone having similar behavior could look at alternative.
Tesla1856
8 Wizard
8 Wizard
•
17K Posts
0
February 9th, 2017 21:00
They just say that as shipped, with various "validated cards" ... it works. I think that list is different for each x51 Release (R1-R3).
Maybe some X51 330w Power-Supplies (combo of AC-Adapter and Power-Board) are weaker than others. Possibly even worn-down over time ... running them near 100% much of the time.