Jump to content

Axymeus

Member
  • Posts

    19
  • Joined

  • Last visited

Awards

This user doesn't have any awards

Axymeus's Achievements

  1. None at all. The system seems to be performing adequately, except for inconsistent frame times in games, causing severe micro-stutters. At least I assume this is the source of the issue. In a 60-fps game, frame time varies from 9 to 20 ms, deviating from the target 16 ms by 30% on average. I can enable v-sync to get rid of the issue on the monitor, but the stutters remain clearly visible in any recordings where I am unable to maintain a smooth 60-fps capture, although no frames are dropped, and both CPU and GPU are at less than 10% utilization. Power settings are set to performance mode too.
  2. Hey, hope this is the right place for this. In an effort to get rid of micro-stutters and high frame times I've tried running LatencyMon. Everything is in the green, but I'm interested in better understanding the result, specifically and generally. I'm open to any resources that could help to better frame the different numbers in terms of scale and decide if they're a non-issue or not. I wonder if they are just right under the threshold for LatencyMon to warn about, but still too high for a stable system. Here are a few questions, for the takers. - Hard pagefaults. This is a RAM access error if I'm not mistaken? In a 10 minutes session process svchost.exe shows almost 1.5K of those, making up for more than half the total pagefaults (2.8K). I seem to recall that pagefaults are not a big issue if they are few of them, but this seems like a way too large number. Is it? - ISRs. 466K ISRs in 10 minutes is once again looking like a giant number, especially since they are all coming from wdf01000.sys "Kernel Mode Driver Framework Runtime" with an additional 46K from HDAudBus.sys. Highest ISR execution time is a mere 144 µs so I think it's fine, but it's still a lot of interruptions on an idling system. Is this expected? - DPCs. I am less familiar with those… They can be seen spread over a fair number of drivers with afd.sys, nvlddmkm.sys, and wdf01000.sys taking the lead, with a total of 3M. Highest execution at 380 µs. Same question as before, is this a standard behavior, should it be lower? For the record, system is running Windows 10, fresh install, all drivers up-to-date. Hardware consists of an i7-11700K and a RTX 3070. Thank you.
  3. True. But in the end nothing beats your own eyes, and they can decide that one monitor is better despite being "objectively" worse than another model. I really wish there was a store near me with all those monitors instead of the generic cheap ones...
  4. Hmm the 27GL950 does look good on paper, and it seems like reviews should be in soon too, promising! However I cannot find a LTT video about it? Only for the 850 (which is pretty much exactly the 83A it seems). I agree that contrast on a IPS is difficult, but my UD58 looks much better in that area even though it's an IPS. Maybe it's not about contrast though, it's really hard to tell. But overall the color look so much better than the GL83A... According to rtings the difference in contrast is marginal 989:1 vs 894:1 but they are pretty comparable in other areas so I assume this small boost in contrast is what makes it better.
  5. Well I hope this won't be bothering anyone as much the title suggests I am in tough spot when it comes to my monitor transition. I am currently sporting a LG 27UD58-B (27'' 2160p 60Hz IPS) that I am quite happy with, except for the 60 Hz limitation. I am faced with an issue that apparently more and more people are experiencing where V-Sync is causing stuttering in games. I'm still not sure where this is coming from, could be a nVidia driver bug, or a Windows bug, or the fact that the monitor is 59.94 Hz and not true 60, thus loosing sync for one frame after a couple cycles... makes no matter. I've tried everything, even changing a lot of parts to no avail. So I've come to the conclusion that my best bet was to switch to a G-Sync compatible 120+ Hz monitor. This train of thought led me to the 27GL83A-B and... sure my issue was gone, but I had a hard time coming back to a 1440p resolution. Text elements looked so blurry... I feel like now that I have experienced the crispness of 4K I can never go back. Plus the 27GL83A-B has poor contrast in my opinion, colors appeared washed out in comparison to the 27UD58-B. So now, I'm a little stuck and looking for some advice. I'm ready to put down as much as 1000€ if the monitor checks all the boxes, but according to the reviews I've read all 1440p 120+ Hz monitors only have average contrast and I'm afraid they will not be any better than the 27GL83A-B. 2160p 120+ Hz does seem appealing but a good one costs closer to 1500€ as far as I can tell. What could you recommend?
  6. Only the one. You're right it does look like I could setup another fan above it I'll have to look into it and get a splitter. There's a free M.2 Fan connector though, I suppose I should be able to use it.
  7. To be clear, this is the setup I'm running I really don't see how I could improve the airflow here. The case could support two 120mm fans at the top I guess but plugging them is going to be an issue. I've just noticed that even at max load my 750W Corsair PSU fan is not spinning though. That could be an issue (although I haven't noticed any power-related problems per se)
  8. Delidding is not something I'd willing to risk. The VRM fan is an interesting thought, the mobo supports it actually so it should be fairly easy to set up, but could it influence my CPU temp? I've installed MSI Afterburner and switched my GPU fans from semi-active to active, now ensuring they spin at 50% all the time, at the very least. They're still remarkably silent, and the GPU temp has lowered to a low 40C, almost 20C of improvement. Because of that the backplate is much cooler to the touch, and the CPU runs 10C cooler.
  9. Fans are running as fast as they can. But the temps are still high, now reaching 100C at max CPU load. The case has an additional fan at the front and at the back, I've checked the air flow and they're all aligned. I guess I can try to screw the heatsink tighter for better contact, but I have no other lead.
  10. You're right I forgot to mention that. I'm using Hardware Monitor for my temperatures. I was expecting my ASUS mobo to come with a monitoring tool but I can't locate it (Armoury Crate is a total mess). They seem fairly accurate to me, but like I said the air flow itself is pretty cool. . The mounting pressure seems OK, I made sure not to overscrew anything. How could I check it? CPU fans are running ~700 RPM on idle they were tuned using ASUS Q-Fan's auto mode. The resulting curve looks good enough but I'll give it a try.
  11. Case is a Corsair Carbide 275R that right now isn't even closed. At 60C the GPU fans aren't even active. Should I go for a custom fan curve?
  12. I've done upgrading my computer, and I'm getting fairly good performance out of it. CPU has not been upgraded so I was expecting it to be the bottleneck, although the passmark results seem a little on the low side, what do you think? Intel i7-8700 Asus ROG Strix Z390-E Gaming MSI RTX 2070S Gaming X Corsair Dominator Platinum DDR4 16Go 2666MHz PassMark results are Memory=3384, 3D=21332, 2D=1122, CPU=13313 (global rating 7281). Is a 70th percentile score acceptable for a 8700? (I was hoping to upgrade to 9700K later on). What really troubles me are my CPU temps. Idle temps are around 45C and when CPU load reaches 100% it climbs up to 90C. I don't think these temps are a real problem, but I don't know how to cool it better. I have a Noctua NH-U14S with an additional fan (one pushing, one pulling). I've checked my thermal paste multiple times and everything seems in order. Only thing is, if I dismount the Noctua after a few run it looks like the paste has spread towards the edge of the heatsink leaving not enough in the middle. However, when I apply paste, place the Noctua, and take it back, the spread is perfect. I've even tried two different pastes (Noctua, and Arctic). Do I have a problem here? I'm also wondering if the backplate of the RTX (very hot to the touch, GPU temp 60C on idle) does not mess with the CPU sensor, since heat is propagating upward, right under the heatsink. Because, even under load, the air coming out of the CPU fan isn't much hotter that room temperature (around 35C these days). I don't know if all of that made sense.
  13. Fair enough. But I hope the Corsair HX750 is quality enough. Of course, it could be faulty but... Well I just want to know before attempting to swap either the mobo or the PSU.
  14. PSU would be a surprise. 750W seems far enough for even the 2070. Considering the 1060 draws less and still exhibits the issue I doubt this would be the culprit. Especially since the PassMark scores are so fine. On the other hand, changing the mobo is going to be a PITA... No test I could run to find that out?
×