/tags/benchmark/Recent content in benchmark on Neolisk's Tech BlogHugo -- gohugo.ioenneolisk@gmail.com (Victor Zakharov)neolisk@gmail.com (Victor Zakharov)©2020-2024 Victor Zakharov. All Rights ReservedSat, 24 Oct 2020 00:00:00 +0000/posts/2020-10-24-xpg-gammix-s11-pro-ssd-benchmark/Sat, 24 Oct 2020 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:13:30 -0400/posts/2020-10-24-xpg-gammix-s11-pro-ssd-benchmark/I recently bought a new 2TB SSD to handle my virtual machines. This one is m.2, was previously using 256 GB SATA drive. Nothing against SATA, it just can’t compete on speed with m.2. Below is a screenshot from an SSD speed test using Crystal Disk Mark version 6.0.2 x64 (front) with temperature check in Hard Disk Sentinel in the background. 58C under 100% load for a few minutes is not bad on stock cooling.Victor Zakharovhardwarebenchmark/posts/2014-12-20-32gb-adata-uv131-usb3-perf/Sat, 20 Dec 2014 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Mon, 29 Mar 2021 12:24:08 -0400/posts/2014-12-20-32gb-adata-uv131-usb3-perf/This is a well-built metal case unibody flash drive (rugged zinc alloy, according to the manufacturer). It was actually slightly smaller than it looked on the pictures. I got a 32GB version for less than 15 CAD before tax, with shipping included. It’s not the fastest flash drive I’ve seen on the market, but for the money, I think it’s a good investment. Here are the numbers measured with CrystalDiskMark, USB2 and USB3 respectively:Victor Zakharovhardwarebenchmark/posts/2014-01-20-windows-7-performance-host-guest/Mon, 20 Jan 2014 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 05 Jun 2021 11:36:56 -0400/posts/2014-01-20-windows-7-performance-host-guest/I am running Windows 7 x64 both as a host and a guest VM using VMWare Workstation v9. Here are the figures of Windows Experience Index for each instance, so you can estimate virtualization efficiency: Host Windows 7: Guest Windows 7 - VMWare Workstation v.9: Guest VM is allocated 2 cores, out of 4 cores total on my Intel i7-4770. As you probably have heard, Windows 7 uses logarithmic scale for its performance score.Victor Zakharovwindowsvirtualizationbenchmark/posts/2012-09-24-dl380-g7-performance-xeon-e5649/Mon, 24 Sep 2012 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2012-09-24-dl380-g7-performance-xeon-e5649/First benchmarked using latest Passmark: Then did some tests with latest Aida64: Memory Read: 2884 MHz - Triple DDR3-1311 - 9-9-9-24 CR1 - 11394 MB/s Memory Write: 2500 MHz - Triple DDR3-1314 - 9-9-9-24 CR1 - 7739 MB/s Memory Copy: 2500 MHz - Triple DDR3-1314 - 9-9-9-24 CR1 - 8489 MB/s Memory Latency: 2900 MHz - Triple DDR3-1312 - 9-9-9-24 CR1 - 77.6 ns CPU Queen: 2500 MHz - Triple DDR3-1314 - 9-9-9-24 CR1 - 64097Victor Zakharovhardwarebenchmark/posts/2012-09-18-mushkin-ssd-hp-dl380-g7/Tue, 18 Sep 2012 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 05 Jun 2021 11:36:56 -0400/posts/2012-09-18-mushkin-ssd-hp-dl380-g7/Hardware: HP ProLiant DL380 G7 Mushkin Enhanced Chronos SSD - 2.5" 240GB SATA III MLC (MKNSSDCR240GB) (RAID1) 2 x Seagate Savvio 10K.5 hard drive - 300 GB 2.5" 6Gb/s SAS 10000 rpm 64 MB Buffer (ST9300605SS) Benchmarked with CrystalDiskMark 3.0.1 x64 on Windows 2008 R2 Std + SP1.Victor Zakharovhardwarebenchmarkssd/posts/2012-05-17-32gb-adata-s102-pro-usb3-perf/Thu, 17 May 2012 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2012-05-17-32gb-adata-s102-pro-usb3-perf/Here are the results from CrystalDiskMark 3.0.1 (x64), connected over USB2.0 and USB3.0 respectively: Very diligent with specs, and actually performing better at USB2 than advertised by ADATA. Looks like a good deal for 30$. Some personal concerns: Neck strap is not included, so might need to buy your own off eBay or similar. Got 5 for 1$, shipping included. Aluminum casing is nice, but the plastic cap does not seem to hold firmly.Victor Zakharovhardwarebenchmark/posts/2011-11-09-acer-desktop-memory-perf/Wed, 09 Nov 2011 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2011-11-09-acer-desktop-memory-perf/Did not have a chance to do a full performance test on this machine (VX4618G-Ei7260W). Memory, however, works quite well for a $700 PC, benchmarked with Aida64: Read: 16.4 GB/s Write: 18.6 GB/s Copy: 18.3 GB/s Latency: 53.1ns CPU Queen: 43.7K (okay, I was just curious). Upgrading stock memory to Kingston Hyper-X 1600MHz CL9 did not improve speeds. It appears the motherboard does not support XMP and continues to run in casual mode.Victor Zakharovhardwarebenchmark/posts/2011-06-02-power-consumption-acer-1810t-notebook/Thu, 02 Jun 2011 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2011-06-02-power-consumption-acer-1810t-notebook/I recently did power consumption test for my Acer 1810T, which I bought a year ago. OS desktop is Windows 7 Home + Aero enabled. I used Kill-A-Watt power meter. The Numbers Off and charging: 21W Idle at desktop + charging: 29W Idle at desktop: 10-11W Half load (one core): 16W Full load: 22-23WVictor Zakharovacernotebookbenchmark/posts/2011-05-17-dotnet-parallel-for-how-good-is-it/Tue, 17 May 2011 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2011-05-17-dotnet-parallel-for-how-good-is-it/I recently had a Microsoft workshop session and became excited about parallel for/foreach/etc. There we got a solid 60% of efficiency. It means that whenever number of cores increases by 2x, performance is boosted by 1.6x. I became curious about it, and created a test project to verify the fact for myself. Here are the results I got on subsequent runs (time in milliseconds + efficiency): I even got 96% on some early build, which did not show that information clearly (hence, no screenshot).Victor Zakharovdotnetbenchmark/posts/2010-12-16-intel-x25-80gb-usage-report/Thu, 16 Dec 2010 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2010-12-16-intel-x25-80gb-usage-report/I ordered mine on 12-Apr-2010, and it’s December now, so it’s ~6 months of light use, 135 days = 4.5 months of uptime and an average 6GB of writes per actual day, which translates in just a bit over 1TB of writes. The drive already has 1 bad sector, which was remapped about a month ago. Decided to test performance with AS SSD Benchmark, and here’s the screenshot with results:Victor Zakharovhardwarebenchmark/posts/2010-07-04-intel-x25-in-hp-proliant-dl380-g5/Sun, 04 Jul 2010 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2010-07-04-intel-x25-in-hp-proliant-dl380-g5/Setup Hardware: HP Proliant DL380 G5, two disks, one is Intel SSD and another one is HP SCSI 15K drive, each running as RAID0. Software: IOMeter, latest version. File system: default NTFS partitions were setup, with Intel being a system drive. It might sound like a bit of overhead (bringing incorrect measurements), but honestly, I don’t think it would really matter on such speeds and IOps. Controller caching: 25% read/75% write.Victor Zakharovdevopshardwarebenchmark/posts/2009-11-15-phenom-x4-9500-vs-phenom-ii-x4-955/Sun, 15 Nov 2009 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Mon, 12 Oct 2020 16:47:55 -0400/posts/2009-11-15-phenom-x4-9500-vs-phenom-ii-x4-955/Introduction My primary concern to upgrade was photo editing. I have a Canon Rebel XSI (450D), which takes 12MP photos. Unfortunately, I’m not yet a very good photographer, probably just need more practice. Anyway, editing photos became a frequent task. For some reason Phenom 9500 is pretty slow on those large ones. And it used to take me around 2-3 seconds to perform simple transformations. Having heard enough about the benefits of Phenom II, I decided to push forward and buy Phenom 955.Victor Zakharovhardwarebenchmark/posts/2009-10-14-win7-on-vintage-pc/Wed, 14 Oct 2009 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2009-10-14-win7-on-vintage-pc/Introduction I recently had a chance to try Windows 7 Enterprise x86 on an old PC. Overall, quite impressed with performance. Tech Specs 1 Pentium 4, 1.6 GHz + 512 MB of RAM on AOpen mobo (dated 2001). Some Benchmarks At startup only 300MB out of total 512 are occupied. The system is clean, no antivirus or anything like that. But running smooth, so it’s still pretty good for 7, compared to its predecessor, Vista.Victor Zakharovdevopshardwarebenchmark/posts/2009-07-07-overclock-phenom-9500-part1-vs-i7/Tue, 07 Jul 2009 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2009-07-07-overclock-phenom-9500-part1-vs-i7/Introduction First, I’m not a big overclocker. Second, I’m not going to use any TEC coolers for ultra fast performance. Third, I don’t have a canister of liquid nitrogen. So, nothing fancy, just air, common non-black processor and a little customized case are going to be involved. If you think it’s pathetic, just don’t read lines below. But if you value your money and want something for cheap, keep going. Still reading?Victor Zakharovhardwarebenchmark/posts/2009-07-02-hp-dl380-g6-benchmarked/Thu, 02 Jul 2009 00:00:00 +0000neolisk@gmail.com (Victor Zakharov)Sat, 24 Oct 2020 12:47:11 -0400/posts/2009-07-02-hp-dl380-g6-benchmarked/Introduction Our company finally bought a powerful HP DL380 G6 server with 2 x Xeon 5540 (8 cores total). Prior to production, we put 2 x 4 gig memory sticks (8 GB of memory) and 3 SCSI hard drives to make RAID5. As you can imagine, initial testing of this box was up to me. Special attention in this article is given to virtualization performance. For us it was primary reason for buying such a powerful box.Victor Zakharovdevopshardwarebenchmark