Jump to content


  • Content Count

  • Joined

Everything posted by Windsurf48

  1. Thanks for the link. I saved the instructions for future reference. I copy the whole hard drive to hot swap hard drives regularly, so I can always restore the system easily. If necessary, I could boot from the backup drive, too. With mechanical drives, backing up or restoring the drive takes about 20 minutes, but now that the system is on M.2 drives, the old SSDs make it possible to backup or restore in under 10 minutes. That enabled me to restore the system to before the update process about 20 times in the past week before I finally got the update applied.
  2. I finally got KB 4532693 to install successfully. I'd downloaded the manual install file, run it, and gotten the same "Preparing Automatic Repair" message. I couldn't get into any of the advanced troubleshooting options because they all required logging on to the administrator account. I was vaguely aware that I hadn't seen an "Administrator" account in Windows 10, but a quick Internet search revealed that it is automatically created but disabled unless you specifically enable it in an administrator CMD window using "net user administrator /enable:yes". After restoring the system, I enabled the administrator account and backed it up again. When I booted back into the problem system using my user account and ran the KB 4532693 manual install file again, everything looked the same going into the restart. I was expecting to find out if I could access the troubleshooting options with the administrator account, but instead the system booted into Windows. I resumed Windows Update and it applied three minor updates successfully, and everything is still working after several reboots. There were only two differences in my final, successful attempt to install KB 4532693: I didn't click 'restart now' but checked the Event Viewer and then used the standard reboot option, and the Administrator account was enabled when the install ran. Neither makes much sense as the reason why the install didn't trash Windows 10 boot files.
  3. This is a different and more common problem: https://www.computerworld.com/article/3528771/with-a-fix-for-the-temporary-profile-bug-still-elusive-win10-1903-and-1909-customers-should-check-p.html This is the problem I'm having: https://www.windowslatest.com/2020/02/16/windows-10-kb4532693-update-is-now-causing-boot-failures/ Unfortunately, the recommended fix is to select Advanced Options and reboot in Safe Mode to uninstall the patch, but my PC won't boot into Safe Mode on either motherboard. The next alternative is System Restore, which is equivalent to my restore from a backup hard drive, but then all Windows Updates have to be blocked indefinitely. I hadn't looked for Microsoft support recently, and found that it is available at $499 per incident. I'd be willing to bet the cost of the incident that what paying for support would get me was instructions to reinstall Windows 10 and see if that helped. Microsoft found another way to make my life more fun yesterday, too, probably to punish me for having motherboard problems. After I swapped motherboards, Office 2016 tried to reactivate and failed. I'd purchased it after Office 2013 said it had reached its reactivation limit last October, and apparently swapping the TRX40 motherboard in and out pushed Office 2016 over the edge. It was sold as having no limit, which seems to be true because it did allow me to reactivate it over the phone, although not before insisting that my cell phone number wasn't a cell phone number so I had to enter the 63-digit number on my phone and then enter the 48-digit response on my PC. Just like your PCs, the update caused no problems on my other Windows 10 system running on the same hardware as the one having problems. There could be some differences in the settings, but most of them are the same on the two systems. I'm going to try repairing the boot files after applying the update, and if that doesn't work, I'll move on to reinstalling Windows 10. If it can apply all the updates, including KB 4532693, I'll finish reinstalling my settings and applications.
  4. I have two hard drives with independent Windows 10 x64 operating systems. One is working fine with all Windows updates as of today, but the other goes into a 'Preparing Automatic Repair' loop every time I let Windows Update try to install KB 4532693. I've restored the hard drive over a dozen times on my Zenith II Extreme motherboard, which is going to ASUS for repair for unrelated issues, and now on my ASUS Maximus X Apex motherboard in the course of narrowing down the problem to KB 4532693, which turns out to have a history of various serious problems. I can pause Windows 10 updates, but then I won't get security updates unless I research, download, and install them manually, or any other Windows updates. Once I turn Windows update back on, KB 4352693 will try to install again and I'll lose the system. I still have few things to try, but I'm not optimistic since KB 4532693 both fails to install and makes the system unbootable, and what I've found searching the Internet is stronger on reporting problems than solutions. Has anyone else been dealing with KB 4532693 problems, preferably successfully? Is there any way to block a single Windows update permanently? Eventually, I'll have to give up and reinstall Windows 10 x64, but that could just end up back at the same point once Windows Update tries to apply KB 4532693 unless I wait until a new Windows ISO comes up with the next feature update.
  5. The Zenith II Extreme is out and I'm back up on the ASUS Maximus X Apex and Core i7-8700k, which are running with no problems. Windows 10 swapped in the Intel drivers with no complications. I'll get the Zenith II Extreme packed up and shipped to ASUS in a day or two. For some reason, ASUS wants just the motherboard returned, not its accessories, and not in its original packaging, so I have to figure out how to get it packed safely for shipment. Recent searches turned up three or four other people having the same problems with the Zenith II Extreme, so I'm not terribly optimistic that ASUS will find and repair problems. That's especially true since ASUS has already replaced the Zenith II Extreme with the Zenith II Extreme Alpha in its TRX40 lineup. If TRX40 motherboards were cheaper, I'd probably just buy a different one and hope that the problems weren't a fundamental part of the TRX40 chipset and/or Threadripper 3960x. If the Corsair RMA process hadn't taken me past the 30-day return limits, I'd just be returning the memory, motherboard, and CPU because they cost too much to make accepting with this many hassles acceptable.
  6. The new power supply arrived this morning and the PC is up and running using it. However, all the same problems seem to be present. So far, I've seen reboots running memory at anything above 3200 MHz and drives disappearing when the system restarts (but never on a cold boot). I'm restoring the other Windows 10 system now to see if it still gets 'Preparing Automatic Repair', but I'm not optimistc. It looks as though the next step will be to contact ASUS about a motherboard RMA.
  7. I just ordered a new power supply. The current one is four years old, which isn't necessarily a problem, but I've had flaky, inconsistent processing which went away with a new power supply several times. I tried restoring the 'Preparing Automatic Repair' system on a different hard drive with a different connection to the system with the same results, so it's not the hard drive. One of the other drives intermittently disappears when the system reboots, so there are multiple issues going on. The memory is running at 2133 MHz, but if it's really bad, it could still be the problem. I'm going to run memory diagnostics against it when I log off, but the original memory passed the Windows diagnostics at DDR4-3600 even though the motherboard rebooted at least once getting started at that speed. I'll try reinstalling Windows 10 for the system which keeps running into trouble if the power supply doesn't resolve the issues, and then contact ASUS support about the motherboard if nothing else makes a difference. Hopefully, Newegg will complete its research into the memory issues with Corsair and decide to give me a break on returning the memory, unless a new power supply resolves the memory issues and that becomes irrelevant. I noticed that several of the newer power supplies needed USB connections to the motherboard for monitoring or RGB lighting. As it is, my case fans use one of the USB headers, and using a second one for the power supply would leave nothing for front panel USB except a single USB-C port.
  8. It looks like the Corsair Dominator DDR4-4266 isn't stable at non-overclocked speeds. It had been running at 3200 MHz, which is the last non-OC value according to Zenith II Extreme specs, apparently without issues, but one of my operating systems began kicking into 'Preparing Automatic Repair' yesterday. I restored the hard drive from a back full copy and the problem went away. The system would boot successfully several times before reverting to 'Preparing Automatic Repair'. I did CHKDSK, SFC, and DISM disk scans, which found no problems after the drive was restored. Neither did Windows Defender, Malicious Software Removal, SuperAntiSpyware, or Microsoft Safety Scanner, but the problem came back. None of the simpler recommended solutions I found had any effect until I found one suggesting that bad memory could be the cause. I've dropped the memory down to 2133 MHz, half its rated speed, and so far everything seems stable.
  9. That's what I used, but I was looking for the fastest 4 x 8 Gig kits available on Newegg, so the selection was more limited. It will be time consuming to research the 3600 MHz kits, but I'll also write to the manufacturer first to find out if they guarantee them at their rated speed. Even if they didn't, one 3600 GHz 4 x 8 kit I checked out cost a fifth of what the Dominator DDR4-4266 cost. I'm not sure what you're paying for if it's only guaranteed to work as well as something costing 80 percent less.
  10. I didn't know about the 3600 MHz threshold when I bought the RAM. However, Corsair wouldn't guarantee its DDR4-3600 at anything over 3200 MHz, either. I tried the RAM I have using Corsair Dominator DDR4-3600 timings and voltages, and that was unstable, too. I'm assuming that there aren't very many motherboards for which anything over 3200 MHz isn't an overclock value, which would seem to give vendors a free ride to charge high amounts for memory rated above that without any responsibility if it doesn't work. Overdrive, AIDA64, Passmark Performance Test, and other benchmarks do show significant increases in RAM benchmarks at 3600 MHz and 4266 MHz compared to 3200 MHz, with most highest at 3600 MHz but some benefiting from 4266 MHz. I doubt that the difference is significant in real world processing, but since my goal was to play around with benchmarks, faster RAM, at least up to 3600 MHz, would be worth it if it actually worked. ASUS bears some responsibility, too, because they provide a QVL list for RAM certified with each motherboard, which would lead one to assume that the listed RAM would work at its rated speed on that motherboard even if it used an overclock value. Unfortunately, ASUS has gotten rid of the individual forums for each motherboard which were the most useful source of tech support from users and ASUS personnel, and its chat, phone, and email support is staffed by low-level people who are very cooperative but end up referring most issues to higher levels, which, based on my experience, may or may not actually have people responding to issues passed to them. At the higher speeds, Corsair is the only company with RAM on the Zenith II Extreme QVL list. However, at 3600 MHz and lower, there are a lot of choices, so if I can return the Corsair RAM, I could start researching a dozen different DDR4-3600 sets, falling back on DDR4-3200 if nothing looked promising. Newegg was at least sympathetic to the fact that I wasted a month dealing with Corsair when I could have returned the RAM within the 30-day window if Corsair had told me up front that they didn't guarantee it to work.
  11. The PC wasn't stable booting with the Corsair DDR4-4266 memory running at anything over 3200 MHz. Corsair RMAed the memory, which took almost three weeks, and the replacement memory has the same issues struggling to boot at any timings and voltages with the clock over 3200 MHz, including the DOCP settings. Now Corsair says that anything over 2800 MHz is overclocked and they don't guarantee that it will work at higher speeds. 3200 MHz is actually the last memory speed not considered overclocked on the Zenith II Extreme, but I would not have paid four times as much for DDR4-4266 as I would have for DDR4-3200 if I'd known that it was only guaranteed to work properly up to 3200 MHz. I've asked about returning the memory but am not optimistic. I believe that I've missed Newegg's window thanks to the three-week RMA, but I see what I can find out.
  12. That's why I've been struggling. I've had a feel for the right amount of paste for the standard-size CPUs, but it's taking some trial and error to get it right with one that big. Because of the shape, a single blob in the center would leave part of the ends without coverage. Spreading an even layer of thermal paste on an area that size is harder, too. I used four small blobs near the corner and one pea-size blob in the center, and that's giving better results than my previous attempts. I'm also not sure what the expected idle temperature is, so it's harder to know whether the paste isn't applied optimally or whether I'm just up against the heat from 24 cores. The best I've been able to achieve so far is 32-35c with either a Noctua U14S TR4-SP3 or Corsair Hydro H150I Pro. It looks as though the Noctua heatsink has slightly better idle temperatures, but the Corsair Hydro AiO may work better at load. However, its circular contact surface leaves a good deal of the CPU heat spreader uncovered.
  13. I'll experiment with the corner and center blob approach.
  14. The pea-sized blob method won't work with the big AMD Threadripper rectangular heat spreaders, although a line down the middle might. Noctua recommended a pattern of nine small blobs and four larger blobs, but that ended up with way too much thermal paste. I may have to try a line down the center, take it off immediately to verify the results, and then try an adjusted amount.
  15. I've been using thermal paste for over 20 years, but in the past month I've run into problems applying two different ones, Thermalright TFX and Thermal Grizzly Kryonaut. Both come with applicators and instructions to spread a thin, even layer of thermal paste over the heat spreader. The problem is that neither paste wants to spread evenly, or I haven't been able to figure out the technique needed to do this. Both want to clump and both want to stick to the applicator more than the heat spreader. In both cases, the heat spreader had been cleaned thoroughly using ArctiCleaner Thermal Material Remover and Surface Purifier, which I've been using for 10 years or so. Most of the thermal pastes I've used in the past had instructions to put a pea-sized drop of paste in the center of the heat spreader or a thin down the middle, but a few have recommended spreading a thin layer of paste and that hasn't been a problem previously. Have thermal pastes become harder to spread evenly? Has anyone used either the Thermalright TFX or Thermal Grizzly Kryonaut successfully?
  16. I've had this with various issues over the years. I had it several months ago on Window 10 running on a GPT drive, and it took trial and error to get the various boot files repaired. It was easier a few years back with MBR drives. If I remember correctly, there've been a half dozen different causes and issues generating the same message. The most recent was something that the May 2019 Windows 10 update did to make Windows unable to boot. I restored the drive several times from full backups and each time the updates ran successfully except for making it impossible to boot once they'd finished.
  17. I replaced the Corsair Hydro H150i with a Noctua NH-U14S TR4-SP3. Idle temperatures are slightly lower, but initial testing indicates that temperatures under load are higher. However, I'm waiting for an RMA for the memory certified for the Zenith II Extreme and using a pair of older memory sticks until it arrives, so there's no point in doing much at this time. The NH-U14S does fit easily on the TRX4 motherboard. The fans have to be raised slightly for taller memory, but with four sticks, the front fan is clear of the memory anyway. AIDA64 shows the fans speeding up under load as expected, so if the load temperatures are still high when I can resume testing, I'll take the heat sink off to verify that the thermal paste was applied correctly and, possibly, try different thermal paste.
  18. The memory has been running at 3600 18-19-19-39 1.35 since noon with no problems, including completing Windows Memory Diagnostics. Unfortunately, I haven't been able to complete Overdrive to see what difference it makes, but the AIDA64 memory scores are encouraging. ASUS's RealBench has similar problems, displaying a message about being unable to parse system specs which blocks uploads after completing the tests. This seems to have been going on for several years with no resolution. It's frustrating because the score looks good compared to the online database, but that might be partly because other high scores wouldn't upload. At least one of SiSoft Sandra's full system benchmark tests chokes on TRX40 systems, but hopefully this is a recent problem which will be fixed soon.
  19. I'm just curious about what kind of rigs are getting the top scores. If they are using LN cooling or are running two Xeons and three video cards, I don't worry about it, but if they are comparable to mine, then it's interesting to try to figure out what's holding mine back. I was trying to figure why the 3960x and TRX40 chipset were getting 10% lower scores than my Core i7-8700k and Z370 in Passmark Performance Test with both running at default settings, and found out that temporarily disabling the Meltdown and Spectre patches resulted in scores 25 percent higher. It didn't make a difference in Overdrive, however.
  20. I forgot the mention that I noticed something about the percents on the Ranking tab. The five individual scores showed 4% each, with no score or percent for Internet bandwidth, and the total ranking showed 20%. Is there any possibility that Overdrive is adding the individual percents to get the overall percent instead of calculating it separately? The 4% calculations are suspect, too, because it seems unlikely that the CPU, RAM, hard drive, and video scores would all fall in exactly the same percent range.
  21. I've seen the black Noctua coolers. Unfortunately, the RAM is too high for either the Noctua or be quiet! coolers. I'll just run the PC without overclocking the CPU and play around with other variables until someone other than Enermax comes out with an AiO pump designed for Threadrippers. My i7-8700k and Maximus X Apex was using RAM designed to fit under heatsinks. I haven't used an AiO since my first one leaked and fried my motherboard three or four years ago.
  22. I checked the QVL before ordering the RAM. That's not the problem. TRX40 doesn't have XMP, and the equivalent DOCP only has an option for 4266,. The problem is that selecting DOCP automatically locks the cores at 4.3 Ghz, which is too much for the H150i. It took me a couple of weeks to deal with driver issues in Windows 10 x64 and I reseated the pump with better thermal paste last weekend, so I'm just reaching a point where I'm ready to experiment with RAM settings. I need to check to see if either the Dark Rock or Noctua's TRX40 cooler will clear the RAM. I've had good results with Noctua heatsinks in the past; the only downside is Noctua's preference for really ugly fan colors. I'm not sure I want to overclock badly enough to set up a custom water loop.
  23. I checked the link for Threadripper coolers. The highest-rated one is the Thermaltake Water 3.0, but it has the same size pump and contact surface as the H150i. The be quiet! Dark Rock Pro TR4 is the highest rated air cooler and has a base which looks as though it would cover the whole CPU, but I don't know if the better coverage with air cooling would be an improvement over the H150i. I'm certainly going to watch for an AIO designed for Threadrippers, but the article says that the one currently available is too unreliable to be useful. I'll also see if I can find any Threadripper 3960x and 3970x stats comparing coolers which include the Corsair H150i and the Dark Rock Pro, but I think I remember seeing some where the liquid cooling trumped the better base.
  24. I've tried the RAM at 3600 MHz, but the system is not stable. I probably need to increase the power, which defaults to 1.2 but goes up as high as 1.45 for 4266 MHz. That's one of the things I haven't gotten around to trying yet. Thanks for the link to CPU coolers. I already had the H150i and some of the 3960x reviews were done using the same cooler, but I assumed that some of the heat issues were related to the size of the pump's contact area and was planning on looking for one designed for Threadrippers. By the way, I completed another run today on my other hard drive's Windows 10 system and it reported that my 85708 score was in the top 20%, with each of the individual scores in the top 4% except for Internet, which had "soon!" instead of a score. My last previous score was 45583 from 2017, and that was in the top 5% which was expected because it would have been my Core i7-6700k or 8700k. What kind of systems are getting the top scores these days? http://www.pcpitstop.com/betapit/sec.asp?conid=26073290&report=Summary
  • Create New...