Userbench fails Sphere graphics test. Possible solution.

If you are experiencing an error running the userbench suite of tests, and the error indicates “ERROR: Sphere GPU benchmark failed, aborting further tests on this GPU” then this MAY solve your issue.

I isolated this error on my system to the Nahimic service that is a part of the Realtek Audio driver package. To verify this as the cause on your system:

  1. Open the control panel.
  2. Administrative Tools applet. *You may need to change the view in your control panel to large icons to see this as an icon. Check the drop down menu at the top right area of your control panel.*
  3. Click on Services.
  4. Disable the Nahimic service by right clicking on it in the list, properties, then disable. Reboot for that change to take effect.
  5. Test userbench.

IF userbench now runs successfully, then Nahimic service IS the cause. The real solution is to update the Nahimic service to the latest version. In my case I have an ASUS Z370-E Motherboard and there is an update package on the ASUS Support page. IF you have a Z370-E here is the link for the audio driver update. You probably will need to find the update that corresponds to your motherboard at the manufacturer’s web site of YOUR motherboard.


  1. Download the update.
  2. Extract or unzip or unpack the update.
  3. Run the Setup.exe file.
  4. Say yes to the prompt to uninstall your existing driver and install the new one.
  5. Reboot when it suggests.
  6. Wait for the installer to resume after the reboot.
  7. Reboot again after the install completes.
  8. Run the AsusSetup.exe to update the Nahimic service, this won’t work correctly if the service is disabled. Enable it and reboot if necessary. Reboot again AFTER the AsusSetup.exe completes it’s install.
  9. Test userbench with the new version of the Nahimic service running.

Your mileage MAY vary. Good Luck!

The Photo FINISH

Boot time variance between two high-end rigs

I have two computers that I use for my primary desktops. One I built in July of 2018. The other I built in April of 2019.

System 1

Intel i7 8700K 4.3Ghz CPU
ASUS ROG Strix Z370-E Motherboard
Corsair Carbide 600Q inverse Case
Corsair H115i Pro CPU Cooler
Corsair HX750i Power Supply
G.Skill Ripjaws V DD4 32G RAM
Samsung 960 Pro 512G NVMe drive
Western Digital 2TB Black Hard drive
NZXT internal USB USB Hub
Netgear A7000 wireless External wireless adapter

System 2

AMD Ryzen 2 2700 4.1Ghz CPU
ASUS ROG STRIX B450-F Motherboard
Corsair Carbide 100R Case
AMD Wraith Prism CPU Cooler
EVGA 500W Power Supply
EVGA DDR4 2400 16G RAM
Crucial P1 1TB NVMe drive
Netgear A7000 wireless External wireless adapter


Systems were both rebooted 10 times with Windows 10 Pro with the latest updates installed. The times were averaged.

System 1 reboots in 23.47 seconds

System 2 reboots in 36.59 seconds

Video performance: Intel Core 2 Quad QX9650 vs Intel Core i5 660.

Lightsmark Benchmark Result i5 660 processor.
Lightsmark Benchmark Result QX9650 processor.

In both of the above benchmarks the following items are identical:

Fedora 12×64

Nvidia Video driver

Evga GTX260 core 216 GPU

4G of RAM

Intel X25-M SSD

The following items are different:

DDR2 RAM for the QX9650 benchmark.

DDR3 RAM for the i5 660 benchmark.

The QX9650 is using an Evga 780i FTW motheroboard.

The i5 660 is using an ASUS P7H55-M Pro motherboard.

The differences in the benchmark results are negligible.   The i5 660 Processor is able to perform within one half of one percent of the QX9650.

Provided the Intel Core 2 Extreme Edition QX9650 processor is an older processor.  But it is a processor that was one of the fastest Intel had ever created 2 years ago when it released.


If you are in the market to buy or build a new computer, the Intel i5 660 makes a very compelling argument for high performance.  Of all things, it is also very energy efficient.   The processor tends to run very cool even under duress.   My testing shows a top temperature of 35 degrees Celsius with stock cooling in a CoolerMaster 690 II advanced case with the stock fans.  If you are using an older processor that is not as powerful as the QX9650 you should see significant performance gain.   Bottom line.  This system has more bang for your proverbial dollar than any other system you can buy or build.   I build and sell i5 660 based systems starting at $1195 plus shipping.  The Intel Core 2 Extreme Edition QX9650 processor cost $1000 on sale typically.


The Intel H55 chipset does NOT support dual video cards OR RAID.   If you are interested in those features you will need to step up to an Intel P55 chipset based motherboard to enable them.  The extreme performance crowd will want SLi or Crossfire support that this system can NOT provide.   Even if you use a P55 chipset motherboard you are limited to two video cards running at 8x performance each.  To obtain the full x16  performance on multiple video cards simultaneously  you have to step up to an X58 socket 1366 motherboard and processor.

Tom’s Hardware benchmark confirms i5 660, i5 661, i7 870 faster than i7 980x.

This benchmark was for Lame and the results indicate that the i5 660, i5 661, and the i7 870 all tied for first when encoding audio from wav format to mp3 format. They all outperformed the i7 980x. Lame is obviously not SMP for more than 2 cores.

I said something to this effect back in September of 2009.
Long before Tom’s Hardware and/or Nvidia mysteriously came to the same independent conclusion on their own. Sometimes emulation is indeed the best form of flattery.

VelociRaptor 10,000 RPM 600G hard drive released!

Today I received the Official Western Digital announcement for the new VelociRaptor 10,000 RPM 600G SATA 6G/s drive. Sounds impressive. If we were limited to mechanical drives, this one definitely would be the one to have for performance. The problem is, we are not limited to mechanical drives. Solid state drives are readily available. The Intel X25-M produces much faster results using a SATA 3G/s interface than the new VelociRaptor drive does using a 6G/s interface. Mechanical drives are wonderful for economical long term storage. The VelociRaptor is far from economical. The 300G model is selling for approximately $200 on Amazon. When compared to a 1TB Western Digital OEM Caviar Black at $100, that is a fairly hefty price for the performance. I recommend using a mechanical hard drive as your “long term” storage of choice. I also recommend using an SSD for your operating system, and key performance hungry applications. ANY SSD will outperform the fastest VelociRaptor any day of the week in any I/O performance test.

Comparison Figures


Intel X25-m
Read 65 micro seconds Decimal notation .000065
Write 85 micro seconds Decimal notation .000085

Read 3.6 milli seconds Decimal notation .0036
Write 4.2 milli seconds Decimal notation .0042

So for read seek times, the Intel SSD X25-m is 55X faster than the VelociRaptor.
For Write seek times the Intel SSD X25-m is 49X faster than the VelociRaptor.

Given that the cost of the drives is about the same the only other issue is space.
The Intel X25-m has 80G
The VelociRaptor has 300G

VelociRaptor has 4X the space for the price.

For my use, I’ll take 1/4 the space at 50X the speed for the same price. Your results and preferences may vary.

My criterion for valid benchmarks.

Recently I have fielded many inquiries related to building the fastest servers available.

So, in reply. I am publishing the list of criterion I adhere to. These are requirements for the server systems that I build and sell for production web or file servers. My standards of quality and moral values in this regard will not waiver based on the lure of easy wealth or time constraints. If I do not have the time to focus the requisite attention on building a server right, it will not be built. I will not cut corners on quality. I build each server by hand. I build each system in Long Beach, California. All sales are direct from If you find a system for sale with my name on it elsewhere, it’s either used (not likely), or an imitation.

1. Stock and reasonably available production parts. Using prototype parts is out of the question. Engineering samples are legitimate if you can later reproduce the result with a production part.
2. Standard clock speeds. Overclocking may be a wonderful option for a gaming system. When you plan to use the system as a production server, that has to be online and operational 24×7 365 days of the year ( or as close to that as possible ) overclocking is not an option. Not to mention that it voids your warranties for the processor and motherboard in most cases.
3. A reproducible production grade build of the system. If you spend 25 hours modifying the system and can not reproduce that with every server you manufacture for sale to your clients, it does not apply.
4. An off the shelf operating system. If you modify the operating system for the benchmark results, and you can not or will not do so for your production systems, it would not apply. If it is a tuning or selection of components that you can and do reproduce for a production system, that’s perfectly acceptable. See rule 2.
5. Reproducible results of the benchmark. If you perform the benchmark once and can not reproduce the results using the same criterion again, using the same production grade system, it is not a valid benchmark.
6. Test tools that are readily available and produce consistent results. I personally prefer Phoronix Test Suite. The tests are standardized and produce consistent results for the same versions of the software. There are other test suites available that provide consistent reliable benchmark results.
7. A system that is configured correctly and securely for a production server environment. SELinux or equivalent enabled in enforcing targeted mode at the very least.

All that said, when following the above guidelines, I build the fastest servers that exist. Faster than Dell 2, 4, 8 or even 16 core servers. Faster than Fujitsu-Siemens Dual Xeons. Faster than even HP Proliant servers. I do not build the least expensive servers available, only the best.

For a listing of the results, the best, the worst, and the mediocre, of my benchmarking results visit:
Phoronix Results

I am never the fastest for every benchmark. I am usually the fastest for Super-Pi and Apache-Build benchmarks. I do get bested on occasion. Currently I am the fastest for Super-Pi to 1 Million digits using the criterion stated above with results that are consistently reproducible and published at Phoronix. Dealer inquiries are not welcome here. Comments and your business are! : )