Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

tarfeef101

Member
  • Content Count

    1,737
  • Joined

  • Last visited

Everything posted by tarfeef101

  1. I've seen some prebuilts from the likes of dell that also use custom GPU designs that allow it to be screwed into the case from the opposite edge of the PCB as the I/O plate, basically locking it into place. that was what i was hoping for, maybe something that would allow me to take a screw out of the backplate of my GPU and put a plate in the middle of it, and screw that into the case to brace it, or something like that. Admittedly a longshot, but figured I'd try. Also i don't think I could, in good conscience, ship something with blue plastic anything in 2020. Even if my friend said he
  2. yeah we discussed it as a backup option, but they're REALLY anti touching electronics. definitely leaning towards that over packing it in and hoping, though. i've had to ship my PC with a courier twice, and both times it dislodged, despite fairly tight packaging. it was a pretty heavy system since it was watercooled (custom, so lots of liquid vs AIOs), which didn't help I'm sure. but still definitely made me wary
  3. yes, i know. but as alluded to in the question (and even implied by watching the latest episodes of secret shopper), they are not foolproof, and are a common source of error even for SIs who use them. If there is an option that's just better, I'd at least like to know it exists
  4. Hey all I'm building a PC for a friend who I've always built and upgraded for. However, we are no longer in the same area so I can't just drive over to his place and give him the PC. So, I was wondering if anyone knew of any cases that have a built-in mechanism to secure a GPU down during shipping (like some OEMs do for their prebuilt systems), since they are very unlikely to be willing to open it up and reseat a GPU if it comes loose (which is still very much a risk even if one uses expanding air packets, etc). It's probably a bit of a long shot, since that's a pretty niche use case for
  5. I find the 100% all the time nature of it suspicious though. That should be throttleable. Especially since my GPUs in every system I've ever used have been pegged at 100% usage regardless of the cpu being used, even if they have wildly varying single threaded performance. I think the cpu resources are being over consumed. I'd hoped ``` <cpu-usage v='10'/> ``` would have been enough to keep the load more reasonable, but it has done nothing
  6. Big oof, my server seems to be crashing after some time running this for whatever reason... worked fine overnight last night, it's been able to train neural nets, mine crypto... not sure what's causing the failures, and don't have time to investigate. guess it's sweater time til I can dig into this (actually can't turn on heating, so folding is my heat in the winter)
  7. Dang. that sucks, my CPU is much warmer than it needs to be now and the fans are louder as a result. got any idea how to make those limit variables actually work? i set the CPU core limit to 10% and GPU to 80%, but both are pegged at 100% still
  8. on my config, ryzen master seems to report accurately, though i haven't tried to modify settings with it as i don't do software OCing (multi-os user here, which makes that pretty difficult/impossible). try wiping it, getting your newest chipset drivers, bios, and ryzen master, and only open it, but don't change settings or modes?
  9. idk what the BIOSes look like on Asus, but when I got my 5950x on a b550 vision D, the BIOS it shipped with was old, didn't report clock speed, and detected half of my RAM. updating DID solve that. does the clock speed show in BIOS?
  10. I'm trying to use the `fah-gpu` docker container to get some extra folding done. I managed to get it to run, use the GPU, allow connection to fahcontrol on my desktop, etc. That's all great. The issue I'm running into right now is that I only created 1 slot, for my GPU (which was intentional). However, it seems that 1 core is being pinned at 100%, despite me explicitly (trying to) configure it not to use the CPU. Anyone know how I can avoid this happening? I've included my `config.xml` below <config> <!-- Folding Core --> <gpu-usage v='80'/> <!-- Folding
  11. I was looking at the 2 R9s, so 0 stock online at any point for any canadian retailer (with the exception of a bug on the memex site). I tried at 6am PST to buy online, failed miserably, and went to line up after that. Was in line for a bit over 3 hours. 4th in line. Didn't get the single 5950x in the store, backordered one. Got the 5900x they had, though.
  12. Yeah you're fine, you can update it with just power and a USB key
  13. Probably more, because more people can afford it.
  14. Yeah that is probably fake, you're right. Look for a used RX 570 if you just want something to tide you over until you get your 5700XT back, or upgrade to a 3080. They are cheap, and won't lose much value if you want to resell it in the next couple months since new cards aren't in that price or performance category. e.g. https://www.ebay.co.uk/sch/i.html?_from=R40&_trksid=p2334524.m570.l1313&_nkw=rx+570&_sacat=0&LH_TitleDesc=0&_sop=15&_osacat=0&_odkw=rx+580&LH_BIN=1
  15. Well too late for the event this year, only just got my setup set up at my new apt. But I'll definitely be folding more now, I have electric heating, and I already pushed wooden desks right against the baseboards so this is my only choice to heat up the apartment Good luck to everyone participating in the event!
  16. W/ respect to RAM, they're basically fine. Only a few manufacturers of the actual DRAM chips, and that makes it pretty hard to mess them up. I'd avoid them for super high speed stuff like DDR4 4266+ since at that point, having your motherboard support your sticks on the QVL is pretty helpful, and bigger brands are more likely to end up on those. More broadly, yeah, it totally depends. PSUs, I'd typically avoid. Watercooling parts, most of mine are from AliExpress. It really does depend. I gave the RAM answer based on the brands you gave.
  17. Well unless you're measuring that with an oscilloscope, kinda. Onboard sensors aren't famous for being the most accurate. Some chips/boards expose die sense voltage, but that has to be done and your monitoring software also must be reading that sensor and not another. Also unless you manually disabled all forms of boost on the CPU, it very well may still boost over your settings in the bios
  18. Yeah that's normal if you're not under a heavy multi-core load. Idle and single core goes pretty high relative to all core by default, which is fine. Don't be worried about it
  19. Well, idk why you're not trying to run at 3200. That'll help way more than better timings at 2400. A rough guideline is that your timings are "good" if the primary CAS latency results in the following inequality being true: Speed (MHz) / CAS > 225 To be clear, if it's worse than that, that doesn't make it BAD, it just not impressive or anything. Some of the best value kits out there are 3200 CL16 or 3600 cl 18. Those would result in values of 200 in that equation. Both are totally fine, just not "good". If you're even lower than 200, I'd consider that to
  20. That depends entirely on the RAM speed, capacity, number of DIMMs, motherboard/CPU, etc. We need more information.
  21. Just want to shoutout to Thermaltake:
    I moved and lost some case feet and a PSU cable (the latter of which wasn't from thermaltake). 

     

    I asked them to send a replacement, more than 2 yrs after purchase, and for a reason that was totally my fault. Under 24hrs they already sent em out to me. I am impressed, and happy. 

  22. So I have 2 setups from which I think I can draw relevant experience: 1. My personal setup in my tiny apartment: I have a pfsense box (box is a bit of a generous term, it's actually just a motherboard and psu on top of a storage bin) as my router/firewall, and a r7000p in access point mode for my AP. It works great. Pfsense, as one might expect, does everything you might want, and more cause it's just freebsd, has dark mode :), etc. 2. My parents' setup I helped with last time I was there. I left them with a ubiquiti edgerouter x (the cheaper end of their lineup), and a TP-Link EAP AC1
  23. Idk, I've worked in a good few places over north America, and there's always been a pretty hard line between the DevOps team and the normal dev team (and QA, for that matter). Most devs I talk to have no desire to get into the weeds of CI/CD, and want to just be able to run a script someone made for them or just push their code and let a pipeline do all the work to build, test, and deploy.
  24. I brought up the slim versions of those images just as a cool fact, almost, and to provide an example of how Centos/debian/Ubuntu aren't all the same. Even debian slim is orders of magnitude larger than alpine. About man-hours vs cloud costs, I'm not sure why you are having issues there. Sure, there's always niche cases where a development process is different from the norm and you could be that. But generally speaking most devs don't interact with the container itself. They just work on the app code and run their tests, and just assume it works. The few devops people
×