Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by Windsurf48

  1. What I see when I update my profile is different from what is showing as my signature, unless the signature is displaying differently for different people. I still see the old Z170 signature under the message above, and this is what I see in my profile: System Specifications: Mountain Mods U2O-CYO case, horizontal motherboard mount, Asus ROG Maximus X Apex, Intel Core i7-8700k, Deepcool Gamerstorm Assasin II CPU Cooler, 2 x 8 gig Corsair Vengeance DDR4-4500, 2 x Samsung 950 Pro 512 M.2 RAID 0, Samsung 850 Pro 1T SATA, MSI Geforce 1080 ti Duke 11g OC, BenQ PD2710QC monitor, Seasonic Platinum 1200 power supply TechExpress Link: http://www.pcpitstop.com/betapit/sec.asp?conid=25986920
  2. Thanks. I'll try that. Does it change when any of the hardware changes? It definitely changed at some point last week, possibly after I installed a new video card. However, today's run overlaid the one from several days ago, although the difference wasn't significant. I remember testing repeatedly a few years ago when the first quad code CPUs came out, and then the ids changed once or twice a day with the same hardware and computer name.
  3. How often does the id number in the Overdrive results Internet address change? I thought it was daily, but I noticed at least one time when it was the same the next morning as the night before. Is it 24 hours based on the first test of the day, or something else? If I get a good score, I don't want to risk wiping it out of the changes for the next run get a lower score.
  4. I learned something yesterday. Thursday, I ran the HWBOT benchmark, which requires turning on the HPET timer. I left it on and then forgot about it when I was running tests yesterday, getting results as much as 40 percent lower at the same settings. Turning the HPET timer back off brought the scores back to the previous values.
  5. I'm not sure if this is the correct forum, but I updated my profile for changes to my system. The changes show on my profile, but not in the signature at the end of messages, and I haven't managed to find a separate signature area to update.
  6. Is it possible to select the server in the Overdrive Internet speed test the way it is in the stand-alone Internet Speed Test?
  7. I'm happy with the upgrades. There must be some very fast systems taking the test because mine is in the top zero percent for disk, but the top one percent for everything else except Internet downloads (the Albany server barely makes it into double digits for some reason). That implies that there are a number of other results for CPU, memory, and video in the same range or above mine. That's what I'd expect based on the range for other benchmarks, but I'd be interested in knowing what the top scores currently are and what systems are getting them.
  8. http://www.pcpitstop.com/betapit/sec.asp?conid=25986920 5 GHz turbo with aircooling.
  9. I upgraded my video card today and got a better result: http://www.pcpitstop.com/betapit/sec.asp?conid=25986920
  10. I went back over some previous test results, and found that my Core i7-3960x was getting scores in the 44,000 range.
  11. After seeing the expected download speeds from Dallas, San Francisco, and London, but not Albany, I'd come to the conclusion that the problem was something beyond my neighborhood and unrelated to my Comcast service between my area and Albany, which might include New York traffic. All the drivers are up to date, and I tend to check them daily for the first few months after upgrading my system. ASUS just released a new BIOS last week, so I upgraded to that. I don't know if that's why I was finally able to get through the Windows memory test at the rated speed of DDR4-4500, although I guess I could ask ASUS if there were any changes related to memory in the new BIOS.
  12. http://www.pcpitstop.com/betapit/sec.asp?conid=25986342 This is my current score after upgrading to a Core i7-8700k and ASUS ROG Maximus X Apex motherboard. The RAM was slightly underclocked because it had been having stability problems at its full rated XMP speed, although that issue seemed to be resolved when I ran the Windows memory test this afternoon, possibility because of changes in the latest BIOS. I think that the CPU can be pushed a little higher then the 4900 MHz limit used for this test, although probably not for long-running tests like PCMark 10 and SiSoftSandra. However, Intel Extreme Tuning Utility showed no throttling running its benchmark at 5 GHz. The video cards and hard drives are the same as on the previous ASUS Z170-WS and Core i7-6700k, so any differences are from the CPU, motherboard and chipset, and memory.
  13. I just ran four Bandwidth Speedtests using the Albany server. All four showed 12-13% CPU usage and no disk activity. One showed 95 Mbs and the other three showed downloads in the low 20s. Upload speeds only have only varied a couple of Mbs between any of the runs. Test against the Dallas, San Francisco, and London servers got 180 Mbs. A final test going back to the Albany server got 21 Mbs.
  14. I'll run the stand-alone Internet speed test with Task Manager monitoring CPU usage. I want to try that test at various times to see how it varies anyway. However, I've never seen much CPU usage just running downloads, and it seems unlikely that downloads would put much load on a Core i7-8700k (by the way, I've been trying to update my profile, but it always ends up at 0 percent complete after going through all the screens, so I've sent an email asking about that). In the past, I've seen some variation in PCPitstop download rates, but not as wide as the current ones. I've seen consistent download speeds using www.speedtest.net, but that uses a server within 50 miles of my house. So far, it hasn't shown bottlenecks at the times I've gotten really slow downloads in the PCPitstop test, but I've seen radically different results using its different servers. I doubt that it makes a difference, but there is a dying locust tree outside my window leaning against the Comcast wire between two telephone poles. Last year, it was six inches off the cable, but about a third of the base of its trunk has rotted and it's been leaning against the cable since spring. I know it's the Comcast cable because the power company came out four months ago, acknowledged that there was a problem, but said that it wouldn't do anything because its cable was above the top of the dying tree and it was Comcast's problem. The tree is on the county right-of-way, but the county can't touch trees near wires, and Comcast can't remove the tree without permission from the county and power company, so it's probably going to stay the way it is until it falls and takes out the Comcast cable. Merry Christmas.
  15. Two Samsung 950 Pro 512M M.2 drives in RAID 0. The Overdrive scores have been around 2700 MB in all the recent test runs. Yesterday morning, the download speed was about 180 Mbs; last night, it was 15 Mbs, and this morning it's back to 180 Mbs. I don't know if time of day and the Comcast connection is a factor in the variations, but the response to my television remote varies widely, too; typically, the cable box stops responding to it completely for five minutes at 8 pm, 9 pm, and 10 pm, and tends to be slower between 7 pm and 11 pm. Comcast has been no help, mostly claiming that it couldn't be happening or telling me to reset the cable box, which takes longer than waiting for the response to pick up again and doesn't resolve the problem.
  16. There is nothing fancy in the desktop settings except the background changing every 30 minutes. Visual settings are set for best performance, and I change settings from 200% to 100% for tests. The indexing service is turned immediately after installing Windows. The only antivirus running is Windows Defender, although I could try disabling that. However, any of the current settings would have been in place when I was testing a couple of years ago after installing a Core i7-6700k. Tests for previous systems would have been under Windows 8.1 or 7, but still with equivalent settings. However, in all previous testing, the download speeds were consistent in tests run within a few weeks or months, only changing when the cable limit changed. It's only been in the past week or two that I've seen widely varying results, and then only for the download speed. All other system results are stable from run to run, including ping and upload speeds. It is possible that something the system is doing is interfering with the download speed, although Windows Update is set to ask before downloading updates and to check for updates every 22 hours, which should limit its interference, and most of the other phone-home functions are disabled using Group Policy Editor or uninstalled used PowerShell. I am investigating some items revealed by O&O ShutUp10, although some that it says are only partially disabled have been uninstalled. I ran Overdrive twice this morning at different settings which should not have impacted download speeds, and both times the download speeds were close the rated 180 Mbs, so I did not see the problem in either run. I was going to do another run tonight, but the CPU was running hotter than I was comfortable with at load, so I backed out the settings instead.
  17. I just upgraded my CPU and motherboard and have been running Overdrive Tests. I'm finding that the Internet download speed ranges from 12 mps to 180 mps. Running the Speedtest.net test after the slower tests gets 180 mps. I don't know if the problem is something in Windows 10 using bandwidth while the test is running or with the connection to the Albany server.
  18. I did some research on Internet Explorer, and it looks as though there are no separate 32-bit and 64-bit versions in the later releases, with a single version using whichever is appropriate for each web site.
  19. It's the version from C:\Program Files, which at least was the 64-bit edition of IE 9 two days ago on the laptop, but now is the 32-bit version after my company's tech support installed IE 10. I'll need to check the Windows 8.1 Program Files folders again to determine if I've actually been running the 64-bit version or not. I assumed that the 32-bit version would be in C:\Program Files (x86) and the 64-bit version, in C:\Program Files on 64-bit versions of Windows 7 and 8, but it turns out that that is not always true.
  20. My office laptop has issues with Java which turn out to be related to Windows 7 x64 defaulting to 32-bit Internet Explorer, so installing 64-bit Java doesn't actually do anything. I decided to try Overdrive using 32-bit IE instead of 64-bit IE, and I did get complete test scores at stock speeds, although it wasn't very impressive. However, when I rebooted to set the PC to its current stable overclock settings and reran Overclock in 32-bit IE, it still got a negative video score and the message that the video test wasn't run. For what it's worth, the problem is occurring in both 32-bit and 64-bit Internet Explorer.
  21. Actually, I read the full license and I think I remember similar statements about sending information to Microsoft with previous Betas, which is one reason I read the EULA more carefully than I normally do. I assumed that the point of releasing Betas and previews was to get that kind of feedback, and, aside from the risk of running anything critical on an unstable operating system, one would simply not do anything on the Beta/preview that one did not want Microsoft to know about, or anyone else, in the event that there was a Microsoft security breach. For me, it is similar to not worrying about whether my company is monitoring my work email account. The company owns the account, so I don't write anything that I would not want anyone else in the company to know about. If I want to complain about other employees, for example, I do it in conversations, not email. Besides, I worked with someone who used company email for personal communications and accidentally sent a very personal email which included relationship details and criticisms of management to the full department distribution list, so I avoid writing anything which I could accidentally mishandle.
  22. I installed Windows 10 Preview x64 on my PC in a separate primary partition on one of two MBR RAID arrays. Both arrays are copied in full to separate hot swap SATA drives before any major changes, making it possible to restore all or part of the system easily. All data is on logical partitions which are duplicated on each RAID array and synched periodically. Without the ability to restore the full system easily, I would not have installed Windows 10 Preview outside a Virtual Machine, and there is no guarantee that Windows 10 won't trash another system just because it has not trashed mine so far. I've had to restore the system until I figured out how to keep a new Windows install from trashing previous ones in other partitions in the past, but Windows 10 Preview x64 installed cleanly in the selected empty primary partition, without any impact on Windows 8.1 x64 in another primary partition on the same array or Windows 7 x64 in a primary partition on the second array (I switch between Windows 7 and Windows 8.1 by selecting the boot array in the Adaptec BIOS). I loaded the x64 ISO onto a USB 3.0 memory stick and the installation went very quickly. Windows 10 Preview had no drivers for the X99 chipset or Geforce GTX 980, but the Windows 8.1 x64 drivers work without any discernible issues. After installing them, Device Manager shows all devices identified without issues. I installed the Java 8.20 x86 and x64 versions for Windows 8.1, and Java seems to be working correctly. Unlike Windows 8 and the 8.1 upgrade, I did not have to search the Internet on a second computer to figure out how to install Windows without using a Microsoft account; there were no references to Microsoft accounts until I selected the Start Menu option to send feedback to Microsoft. Because I live alone, I don't use a password on the PC, and Windows 10 was willing to install with a blank password and boot into Windows without stopping at the logon screen. Updates go very slowly for some reason but complete successfully, but booting into Windows is fast, about 8 seconds on top of the motherboard and Adaptec 7805 SAS controller times. Overdrive crashes "Checking Windows Setup", which can't be bypassed in the debugging version. GeekBench 3 won't run, either. The current version of SiSoftSandra completes the full system benchmark without any problems although it thinks that it is running on Windows 8.1, and Passmark Performance Test 8 also runs without problems. AIDA64 works, and I ran several Futuremark benchmarks without problems, too. Passmark and Futuremark scores seem to be slightly lower than those under Windows 8.1 x64; SiSoftSandra scores are about the same at stock speeds and overclocked at 4 Ghz. For some reason, the Start Menu Shutdown exits Windows 10 but does not turn off the PC, but a "shutdown.exe /s /t 0" shortcut works correctly. Overall, there is a pleasant lack of major issues or struggles to figure out the interface. On the other hand, there is nothing particularly exciting, either, although it's good to know that I won't be forced into an operating system whose interface I like less after new hardware comes out without Windows 7 drivers. The revived Start Menu is easy to use; but its graphics carried over from the Windows 8 Start Screen still look clunky. At least the colored squares are smaller, but it seems like Microsoft could come up with some more attractive options for PCs with decent-size monitors.
  23. Thanks. I didn't see it as an urgent problem since Windows 10 Preview could change while tests using it were being debugged. The installation process was simpler than past Betas or new versions of Windows on my PC. I was able to install it in a separate primary partition on an MBR RAID array without any impact on Windows 8.1 in another primary partition and without any impact on Windows 7 on a second RAID array (it's booted by switching boot drives in the Adaptec BIOS). There were no issues loading the ISO to a USB drive, booting from it, and selecting and installing into the desired partition. Unlike Windows 8 and 8.1, it was not necessary to do research on the Internet to figure out how to install the operating system without using a Microsoft account. So far, benchmarks have not been encouraging, although not far off Windows 7 and 8, but there are a lot of services and system programs running and it's not easy to tell which ones might be slowing things down. At this point, Windows 10 Preview works and doesn't have any annoying surprises, but doesn't have any exciting novelties or performance improvements, either.
  24. I tried again running in debug mode with everything checked and got the same results. Since "Checking Windows Setup" isn't one of the items skipped, this isn't surprising.
  25. Running Internet Explorer as administrator in Windows 10 Preview, Overdrive gets through the first five lines, reaching "Checking Windows Setup". At that point, a windows pops up stating that Internet Explorer has encountered a problem and must close. IE restarts, gets to the same point, brings up the same window, and shuts down. Java 8.20 x64 and x86 are installed and Overdrive has no problem with them in Windows 7 x64. It's not too surprising that there are glitches with Windows 10 Preview, but this is the first significant one I've come across. AIDA64 works, and so does Passmark Performance Test 8. The Shields Up! port tests work. Installing updates seems to be very slow but work eventually. There were no drivers for the X99 chipset or Geforce GTX 980, so I used the Windows 8.1 x64 drivers. Device Manager shows that all devices are identified and have drivers. There seems to be a minor glitch shutting down Windows, which blanks the screen but leaves the PC running. However, using SHUTDOWN.exe /s /t 0 works as an alternative to the Start Menu shutdown.
  • Create New...