Posting this here, as in the next hour I’m getting back my MONSTA! after having used it for only a fortnight, and want to hear from others how it’s failure or success is working for you guys and gals. Am I forcing too much on it using GPU accel + NEAT VIDEO which uses my GPU to great effect?
I mouth-watering, be aware be murdered!
Huh, your MONSTA rests more in sick-bay than doing it's job. Isn't your MONSTA water-cooled?
If I understand your question correctly, you are asking if you are working your system too hard and asking if that is why multiple components failed & had to be replaced by your builder under warranty...
If the system was assembled correctly with adequate cooling, power supply, etc. you should be able to run it 24/7 365 days per year without any major issues. Ethereum mining rigs run continuously & push the GPUs, especially, much harder than Vegas ever will... Servers, NAS, etc. run continuously...
On my DIY 9900K / VEGA 64 LQ system, I can load (6) different client's lengthy Vegas projects (historical footage for documentaries, etc.), and set them all up to batch-render multiple file types. Sometimes my system will run for 24 hours with the CPU at 100% and the GPU over 50% and when I return to that system, all of my files have rendered correctly. A lot of my footage is from old films so I often setup Mercalli SOL & Neat Video to stabilize, reduce grain, etc. and they can both push the system to 100%... I also have (3) DIY Intel X58 systems with 6-core Xeons & 980x CPUs that have been mostly trouble-free for over 10 years now. I worked those systems especially hard when I first got into 4K & now I keep using them to import from legacy media, like films & tapes, so I can keep editing on my faster system...
What can KILL a system is POWER-related issues. Inconsistent voltage from the wall, power spikes, brown-outs (voltage drops), etc. I have everything of value in my studio connected to various PURE SINE WAVE UPS. My power lines are underground, so we do NOT experience as many power outages as those with overhead lines. Other power issues can be created by a bad power supply. I have been getting the Corsair & EVGA power supplies rated for much higher than what my systems draws, typically 850W to 1000W range...
BTW, the EVGA has a different power pin layout than most of the other Power Supplies although they use the SAME modular power connector. Even the best pro-builders have made the mistake of just swapping-out the power box, and not the cables, only to fry many components attached with the old cables... When you told us so many components of your MONSTA failed, including not showing up until it was in the builder's hands, my gut tells me they did something wrong & are just blaming the components... However, they are making it right by covering everything under warranty - the only problem is that you are now worried you will break it again...
I don't think you had anything to do with it... When you get it back keep it connected to a UPS & run it hard. I'd run some 24/7 stress tests on it while it is still under the new warranty... There is no sense having one of the best Vegas rigs out there & then being afraid to work it...
As to Electricity supply I’ve been running the same setup for twenty years. I’ve got 1kw Corsair.
I mouth-watering, be aware be murdered!
@Yelandkeil - What?
UPS/power regulation are a must. Proper cooling too. All water coolers (is it an All In One?) are not created equal. Adequate ventilation (air flow) through the case, even if the CPU is water cooled is also extremely important. The same with VRM cooling.
I’ve got 1kw Corsair.
I thought you had a 1000W Corsair or EVGA, same as me... Previously I've looked over your build specs and that system is built like a server & should last & last. I really like the stability & longevity of my Asus motherboards. Asus even promotes the Asus "TUF" X299 MK 2 motherboard's 24/7 reliability, military-grade CAPS & MOSFETs, etc. The Corsair memory, Intel CPU & Nvidia GPU are all quality components that last & last running in servers, mining rigs, etc. so if everything was assembled correctly, you shouldn't have to worry.
And since your builder provides a warranty & has to provide labor-time for support, I'm sure they didn't overclock anything & running at safe voltage & clock speeds designed for stability & longevity... However, I overclock all of my CPUs, and the oldest Intels are still running 10 years later... Most modern CPUs automatically throttle-down if they get too hot, so it's hard to kill them...
So bottom line, everything in your system is designed to run 24/7 for at least 5 years. Since you are not using it like that, you should get 10+ years.
@Grazie @TheRhino Some very valid points the Rhino makes about power supplies etc. But then you sometimes have to look at the bigger picture, is there something just staring you in the face that you somehow overlooked?
Of course, the BIOS, software, VP, the weather etc all have to be taken into account. A forensic examination of the smallest detail is sometimes called for.
For example, when after a good nights sleep, you jump out of bed in the morning and head over to your PC, does it sometimes appear even more refreshed than you are?
After all of the above, and still no luck, there's Karma. Have you maybe messed someone around lately? Karma can bite you in the rear, when you least expect it
I’ve got 1kw Corsair.
I thought you had a 1000W Corsair or EVGA, same as me...
So bottom line, everything in your system is designed to run 24/7 for at least 5 years. Since you are not using it like that, you should get 10+ years.
@TheRhino - 1kw is a 1000watts. And yes, I agree.
NOOOoooooooo! Say it isn't so!
Did you tech guy say what was wrong before?
At this point maybe have an electrician check the home wiring? Or else this builder keeps repeating mistakes... Unbelievable!
I would invest in an UPS before turning the computer back on when you get it. Usually, the power supply will stop spikes from hitting the computer. Maybe that is what happened.... and I hope the voltage on the power supply is set correctly. Here in Europe, you can switch them between 110v and 220v....
I am truly sadden by this news....
Anyways, this is all moot - it has failed again after 30 minutes going back Monday. And @JN- if it is Kama, then I’m coming back as a Slug 🐌!
My heart breaks for you... I can't imagine your frustration... Before you send it back, is there someone you know locally that can look it over, look at the voltage going from the PSU to the CPU, GPU, etc., inspect your home electric, etc?
1kw is a 1000watts. And yes, I agree.
Yes, that's what I meant.... You have a quality 1kw /1000W PSU, same as me, so shouldn't have any issues...
At this point maybe have an electrician check the home wiring? Or else this builder keeps repeating mistakes... Unbelievable!
I agree, some home wiring issues can be hidden... After selling & moving out of our last house, I was editing full-time at my wife's grandfather's older home while we were looking for our next house... Although initially I had tested the electric, etc. before setting up my studio, I didn't check it day-to-day. At one point a bad METER provided by the power company began to fail and I was only getting 85-100V vs. 110V (USA) on one leg of the 220V line input... The lights did NOT flicker, the A/C was still running, but my computers would lock-up or shut-down during long renders.
The computers were connected to some older-style, cheaper UPS without digital meters & apparently the voltage had not dropped low-enough to sound their alarms & trigger them to come-on, so the computer PSUs were struggling to maintain 5V & 12V... The power company replaced the meter & wires from the road, and I got some quality UPS with digital meters. All has been good & I am still using those (3) systems 10+ years later for various background tasks, like importing tape media, etc...
OK, the PC is being shipped back on Monday. It’s going to be seen by the Chief Technicians. I’ve already told them to confirm the voltages on the rails. I’ve told them that their tests maybe excellent stress testing for individual items, but they now need to see what happens under actual usage and THEN test the voltages. There is a fatal flaw in the supply of volts which leads onto GPU failure. The chap agrees and is writing a comprehensive report relating everything he witnessed whilst doing a Screen Share session with me. He’s being very understanding and helpful.
Now, where did I put my knitting down? Ho hum......
Feel for you @Grazie . So was it just the gpu that failed? Nvidia 3000-series pulls down allot of juice and although 1000W should do it, you might try bumping up the supply anyway... I think the next step up is 1600W which should eliminate that as a suspect. Fwiw, I've never had an evga power supply come up short on me and they're designed with multiple gpus in mind. Pay particular attention to where the three 3080 8-pin power lines come from... evga supplies provide separate cable connections for 3 video boards and I would suggest pulling a separate 8-pin cable from each. I'm not as familiar with Corsair supplies or whether they can do that.
Btw, what's the motherboard, cpu, ram, etc? Not all brands are created equal and ram timing, matched sets, and density are critically significant... make sure your parts are specifically approved on your mobo maker's qualified vendor/parts list. Also, I recall you mentioned somewhere that you went with 64gb ram... that's risky because that generally doubles ram density per stick for which qualified parts are fewer and timings and speed need to be lowered... but not necessarily an issue if your mobo has enough ram slots to go with 8K dimms.
@Howard-Vigorita - Thanks for the advice. You can read much of my setup in my sig. As to making changes with PSUs and cables it’ll be in the hands of the builders, again.
@Grazie I thought they replaced your PSU when they FIRST noticed multiple components failing... When components with low failure rates, like the CPU AND GPU, both fail at nearly the same time, the FIRST step is to always replace the PSU... I assumed they already did this as it is standard procedure and why I made the following comment back on 2/4/2021:
Sounds strange that you have had BOTH the GPU & CPU fail in the same system... I hope they gave you a quality power supply otherwise you will have more issues in the future...
I... have never had a GPU or CPU fail so having both go in the SAME system is suspicious of something else, like a bad power supply...
Following your Monsta posts, it looks like from 1/20/2021 until at least 4/20/2021, or THREE full months, you have been without a reliable system because they never replaced the original PSU, the most likely culprit... I know you like your builder because they appear to be good people, but sometimes even good people are not always good at what they do... They should have caught your faulty PSU before your MONSTA bled-out & died on the table...