Jump to content

StandardFiend

Member
  • Posts

    12
  • Joined

  • Last visited

Awards

This user doesn't have any awards

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

StandardFiend's Achievements

  1. So, doing that caused my PC to fail to boot. Had to clear the CMOS to get back into the BIOS. I think that means it's time for me to stop messing with this. Without someone who knows what they're doing looking over my shoulder, I'm bound to brick my box sooner or later.
  2. @nick name Here is what's available for CPU configuration with PBO on Manual. PPT (W) is Board Socket Power capability. TDC (A) is Board thermally constrained current delivery capability. EDC (A) is Board electrically constrained current delivery capability. I have no idea what any of that means or what to do with it haha. Also, since I may as well, I'm also including what my voltage settings look like, in case you can see something I've missed.
  3. So I guess the main point is whether or not having the CPU averaging 40-41 (the room is as hot as it's going to get for the year, I believe) is acceptable for the longevity of the unit. @nick name has said that the numbers are okay; do you concur?
  4. Yeah, maybe that's just what life is like with an AIO. Like I said, I've never used one before. I have a custom curve set for my fans, so between 38-40 degrees they are running at 20-30% speed. If I run them at max speed the temps drop a lot, but it's very noisy. Using the Carbide 540 Air. So the radiator is getting a very nice flow of outside air to suck through. The options I have for PBO are Auto, Enabled, Disabled, and Manual. Manual is what gives me the additional menu to adjust settings. When looking at it in the mobo, the settings are another bunch of abbreviations that I don't understand. I'll have another look tonight after work at what the specifics are.
  5. @nick name I will admit that I had to google that to even know what you were talking about. I set it to Enabled (it was previously on Auto) and suffered a further slight loss in multi-threaded performance, although now the cores are clocking .2 MHz higher. I wouldn't have any idea what to do with the Manual PBO settings. I dunno what else to do. Most other posts I've seen in various places indicate that with a double-fan AIO my idle temps should be around 35, and mine are still almost 41. ? But thanks for your continued assistance. I understand that it's probably very tiresome dealing with someone who is a complete noob at this stuff.
  6. @nick name I ran some CPU-Z bench tests for some casual readings, and here is what I got (I ran five tests for each condition and averaged the results): At stock voltages (everything set to auto): Max core speed of individual cores: 4349.0 MHz Max speed of all cores together during bench: 3999.1 MHz Max temp after benches: 60.5 Average temp from boot to cool-down after tests: 43.1 (not a terribly scientific reading) Single core performance (average of 5 tests): 464.6 Multi core performance: 4850.2 With -0.0625v offset on Core and with SoC set to 1.000: Max core speed of individual cores: 4348.7 MHz Max speed of all cores together during bench: 3948.8 MHz Max temp after benches: 57.0 Average temp: 43.9 Single core performance: 460.4 Multi core performance: 4790.8 So, with the voltages turned down, the max boost clock hasn't changed significantly. It still tops out at ~4.35 GHz although for some reason one single core (core 4) tops out at 4323.7 MHz. No idea why that would be. Also, the synchronized max speed dropped by 50 MHz, and while max temp during the tests went down, I didn't see any difference in average temp over time. After a couple hours, it's still sitting at 39.8 average with the lowered voltage. Most importantly, the actual performance has gone down by a few points in both single core and multi core. So essentially all I've done, it seems, is give myself a very slight underclock without any noticeable gains in temperature.
  7. That's a typo on my part, it should have read "under 50%." But you didn't specify how much less. It's at 40 right now.
  8. @nick name Sorry for not tagging, still new here. And thank you for your continued help. With a bit more reading I learned that it's possible HWInfo is slightly under-reporting the SOC voltage. A difference of only 0.025v makes that seem likely. I gave the Vcore an offset of -0.05 as you suggested, but it doesn't seem to have made much of a difference, so I'll probably play with that a little more to see if I can bring the temps down another few degrees. I'm still averaging over 40 during almost nonexistent workloads (average 2%). I set the CPU minimum to 50%, and while the average clock speeds are now accordingly lower, it doesn't seem to have affected the average temp. Minimum is down a degree and a half or so, but the average was unaffected. Maybe I can set the max down to 20-30% for work and see where that gets me. If nothing looks bad, that's good enough for me. I guess that's all I need to know. I'll keep tinkering with the voltage to see if I can drop it, but as long as I can expect a long and healthy life out of the chip the way it is currently, that's all that matters.
  9. Alright, I did that and here are the results (Aida64 stress test and then a couple hours of nothing). I've lost a couple degrees all around. Is this what you would consider normal for my configuration? It still seems a bit high to me, especially for a 240mm AIO. I dunno. How do the voltages look now? Is it normal for the SOC voltage to be sub-1, even though in the BIOS it's exactly 1.00?
  10. Thank you for the reply. Tinkering with voltages (or anything in general) is not something I'm familiar with. Do I do that through the BIOS, or through some other software?
  11. Hey all. My first post here at LTT, with some questions regarding my new PC. It's been a long, long time since I've put a rig together, and there were some firsts for me, most important of which were both the installation of an AIO and the use of an AMD processor. My last machine was a Clarkdale Core i3 dual core using the stock heatsink, for example. Anyway, here is my system breakdown: Momboard: MSI x470 Gaming Pro Carbon CPU: AMD Ryzen 2700x Cooler: Corsair H100i Pro RGB TIM: Noctua NT-H2 RAM: 2x8GB Ballistix Sport AT 3000 Storage: Intel 660p 1TB NVMe GPU: Gigabyte GTX 1660 Ti Windforce OC PSU: Corsair RM650x Case: Corsair Carbide 540 Air The radiator is mounted at the top as an exhaust, with the fans pulling through the fins (which was a recommended configuration in one of LTT's videos). The case has two front 140mm intake fans and one rear 140mm exhaust. Attached are a screenshot of a 10-minute Aida64 stability test, and temp HWinfo readings during the test, during a few hours of day-to-day work activity (which consists of mostly using MS Word and Chrome inside a remote desktop connection, as well as some light Chrome browsing locally), and during a brief gameplay section of Resident Evil 2. During the Aida64 test, the CPU maintained 100% usage and reached a max of 75.6 degrees, and stayed within a few degrees of that for the duration. During day-to-day activity, the temperature averaged 40.3 and spiked to a max of 54.1. I wish there were a way to view median temps, but from what I can gather just by keeping an eye on HWinfo, the processor spends a great deal of time in the low 40s--between 40-44. The lowest temperature for the entire period (about 3 hours) was 35.5. During gameplay it averaged 46.2 and hit a max of 57.6. RE2 keeps a pretty minimal load on the CPU. Ambient temperature of the room throughout the day ranged between 23-24. My radiator fan profile is set to keep the blades at 20% until 38 degrees, when they will start ramping up to 90% at 60 degrees, then 100% at 80 degrees. So during the Aida64 test they were running at 90%. The temps seem to high to me. 35.5 as the low with an average of over 40 for basically doing nothing, averaging 1.6% usage, seems too high. But again, I have no experience with AMD processors and I've never used an AIO before. Which is why I come to you. I'm hoping you can allay my concerns before I dismount the block and see if it's an issue of the TIM not having adequate coverage or if there's something worse going on. I happen to have a thing of Conductonaut lying around. I opted for the Noctua paste instead just for ease of use, but do you think it would be worth using the liquid metal instead? Is there another paste I should reapply instead of the Noctua H2? Thank you in advance for any assessments and advice you can offer.
×