Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jumbosausage]

I've been experiencing issues with JRiver hanging and freezing on the server side. I've been through most of the checks and disabled any anti-virus etc but i'm still seeing strange behaviour. This is running as a Windows 10 VM on a Hyper-v server with the music data being stored on a separate FreeNAS machine. I'm guessing that these benchmark scores are ok but just wanted someone to confirm.


=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 7.159 seconds
    Single-threaded floating point math... 4.460 seconds
    Multi-threaded integer math... 2.506 seconds
    Multi-threaded mixed math... 1.353 seconds
Score: 1228

Running 'Image' benchmark...
    Image creation / destruction... 0.426 seconds
    Flood filling... 0.529 seconds
    Direct copying... 0.481 seconds
    Small renders... 1.705 seconds
    Bilinear rendering... 1.050 seconds
    Bicubic rendering... 0.505 seconds
Score: 4686

Running 'Database' benchmark...
    Create database... 0.210 seconds
    Populate database... 3.253 seconds
    Save database... 1.144 seconds
    Reload database... 0.225 seconds
    Search database... 2.259 seconds
    Sort database... 1.779 seconds
    Group database... 1.481 seconds
Score: 2077

JRMark (version 22.0.88): 2664

[RD James]

AMD Ryzen R7-1700X at 4.05GHz with 3600MT/s DDR4

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 1.734 seconds
    Single-threaded floating point math... 1.858 seconds
    Multi-threaded integer math... 0.436 seconds
    Multi-threaded mixed math... 0.234 seconds
Score: 4458

Running 'Image' benchmark...
    Image creation / destruction... 0.346 seconds
    Flood filling... 0.212 seconds
    Direct copying... 0.461 seconds
    Small renders... 0.915 seconds
    Bilinear rendering... 0.534 seconds
    Bicubic rendering... 0.382 seconds
Score: 7720

Running 'Database' benchmark...
    Create database... 0.176 seconds
    Populate database... 0.993 seconds
    Save database... 0.212 seconds
    Reload database... 0.035 seconds
    Search database... 0.826 seconds
    Sort database... 0.888 seconds
    Group database... 0.536 seconds
Score: 5864

JRMark (version 22.0.97): 6014

[tzr916]

Wondering why a core 2 quad 2.66GHz has a slightly better MATH score than an i5-7500 3.4GHz ?




NEW Server:
Gigabyte GA-H270M-D3H mATX
i5-7500 3.4GHz
16BG PC4-19200
OS drive: ADATA PCIe 128GB NVMe Gen3x4 M.2 (running at 5x SATAIII)
Data drive: WD AV-GP WD30PURX 3TB (running at SATAIII)
GB LAN
NVIDIA GT640
Windows 10 Pro 64-bit

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 4.137 seconds
    Single-threaded floating point math... 2.898 seconds
    Multi-threaded integer math... 2.505 seconds
    Multi-threaded mixed math... 1.710 seconds
Score: 1689

Running 'Image' benchmark...
    Image creation / destruction... 0.210 seconds
    Flood filling... 0.217 seconds
    Direct copying... 0.329 seconds
    Small renders... 1.027 seconds
    Bilinear rendering... 0.793 seconds
    Bicubic rendering... 0.446 seconds
Score: 7281

Running 'Database' benchmark...
    Create database... 0.170 seconds
    Populate database... 0.833 seconds
    Save database... 0.104 seconds
    Reload database... 0.029 seconds
    Search database... 0.870 seconds
    Sort database... 0.795 seconds
    Group database... 0.487 seconds
Score: 6539

JRMark (version 23.0.22): 5170



Old Server:
Abit IP35e
Q9450 Core2 Quad 2.66GHz
8GB PC2-6400
OS drive: Samsung SSD 840 EVO 120GB (running at SATA II due to motherboard)
DATA drive: WD AV-GP WD20EURX 2TB (running at SATA II due to motherboard)
GB LAN
NVIDIA GT640
Windows 10 Pro 64-bit

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 4.359 seconds
    Single-threaded floating point math... 2.589 seconds
    Multi-threaded integer math... 2.580 seconds
    Multi-threaded mixed math... 1.347 seconds
Score: 1747

Running 'Image' benchmark...
    Image creation / destruction... 0.549 seconds
    Flood filling... 1.384 seconds
    Direct copying... 1.949 seconds
    Small renders... 2.177 seconds
    Bilinear rendering... 1.385 seconds
    Bicubic rendering... 0.931 seconds
Score: 2627

Running 'Database' benchmark...
    Create database... 0.420 seconds
    Populate database... 2.295 seconds
    Save database... 0.553 seconds
    Reload database... 0.111 seconds
    Search database... 2.195 seconds
    Sort database... 1.729 seconds
    Group database... 1.049 seconds
Score: 2574

JRMark (version 22.0.110): 2316

[mojave]

Wondering why a core 2 quad 2.66GHz has a slightly better MATH score than an i5-7500 3.4GHz ?

Sometimes the new CPU's run longer at a lower speed before going to max speed. I would monitor the CPU speed during the test to see if that is the issue. If so, just for the benchmark, try disabling all the C states in the bios. You can also disable CPU EIST Function.

[Hendrik]

Instead of disabling things in the BIOS, just set windows to High Performance Power Plan for consistent results.

[defenz0r]

I7 7700k @ 5GHZ AVX = 4.8GHZ
Corsair DDR4 2x8GB 3000 1T
Win 10 x64

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3,056 seconds
    Single-threaded floating point math... 2,104 seconds
    Multi-threaded integer math... 1,014 seconds
    Multi-threaded mixed math... 0,608 seconds
Score: 2801

Running 'Image' benchmark...
    Image creation / destruction... 0,152 seconds
    Flood filling... 0,178 seconds
    Direct copying... 0,313 seconds
    Small renders... 0,767 seconds
    Bilinear rendering... 0,626 seconds
    Bicubic rendering... 0,338 seconds
Score: 9265

Running 'Database' benchmark...
    Create database... 0,129 seconds
    Populate database... 0,711 seconds
    Save database... 0,183 seconds
    Reload database... 0,060 seconds
    Search database... 0,630 seconds
    Sort database... 0,566 seconds
    Group database... 0,407 seconds
Score: 8006

JRMark (version 23.0.22): 6691

[JimH]

I7 7700k @ 5GHZ AVX = 4.8GHZ
Corsair DDR4 2x8GB 3000 1T
Win 10 x64

JRMark (version 23.0.22): 6691

Hot rod!

[Hendrik]

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3,309 seconds
    Single-threaded floating point math... 2,285 seconds
    Multi-threaded integer math... 0,423 seconds
    Multi-threaded mixed math... 0,290 seconds
Score: 3012

Running 'Image' benchmark...
    Image creation / destruction... 0,173 seconds
    Flood filling... 0,249 seconds
    Direct copying... 0,184 seconds
    Small renders... 0,843 seconds
    Bilinear rendering... 0,660 seconds
    Bicubic rendering... 0,349 seconds
Score: 8950

Running 'Database' benchmark...
    Create database... 0,133 seconds
    Populate database... 0,779 seconds
    Save database... 0,110 seconds
    Reload database... 0,049 seconds
    Search database... 0,711 seconds
    Sort database... 0,711 seconds
    Group database... 0,455 seconds
Score: 7294

JRMark (version 23.0.36): 6419

My new workstation, i9-7900X @ 4.6GHz. Unfortunately the MC benchmark is largely single-threaded so no stomping a 5GHz CPU.
But it does build MC in about half the time as the old system.

[RD James]

My new workstation, i9-7900X @ 4.6GHz. Unfortunately the MC benchmark is largely single-threaded so no stomping a 5GHz CPU.
But it does build MC in about half the time as the old system.

That explains a lot. I'm a bit surprised that it's only 5% faster than my Ryzen build, but I guess it's the workload rather than being representative of their relative performance levels. I expect that building MC will be much faster on the i9.
Looking at it in Process Explorer, I was surprised to see how many tasks in MC are only single-threaded, or only scale to a few threads.
SACD decoding for example: I would have thought that 6 channel tracks would at least decode on 6 threads, but it seems limited to 4 at most.
Audio analysis is limited to 8 files at a time. (though decoding an individual file is typically multi-threaded)
Are there plans to improve this, or is that not on the table right now?

[defenz0r]

How it come i get under Windows 7 x64 Prof a better Score?
Any explanation? Its the same latest Version

Even compared with a blank installation its slower

Hell the difference is so gigantic!
Its whopping 8 % difference

In both cases a fresh win installation was used btw.
=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3,041 seconds
    Single-threaded floating point math... 2,100 seconds
    Multi-threaded integer math... 0,979 seconds
    Multi-threaded mixed math... 0,629 seconds
Score: 2815

Running 'Image' benchmark...
    Image creation / destruction... 0,105 seconds
    Flood filling... 0,161 seconds
    Direct copying... 0,291 seconds
    Small renders... 0,759 seconds
    Bilinear rendering... 0,592 seconds
    Bicubic rendering... 0,323 seconds
Score: 9862

Running 'Database' benchmark...
    Create database... 0,125 seconds
    Populate database... 0,587 seconds
    Save database... 0,079 seconds
    Reload database... 0,023 seconds
    Search database... 0,616 seconds
    Sort database... 0,560 seconds
    Group database... 0,382 seconds
Score: 9064

JRMark (version 23.0.36): 7247

[tzr916]

Instead of disabling things in the BIOS, just set windows to High Performance Power Plan for consistent results.

Both systems were already set to Power Plan High Performance. I might try disabling CPU EIST. I seem to remember running that way on my old Q9450 server. Good news the benchmark using 64bit MC23 is better than 32bit, and "Math" average is now virtually equal to the Q9450.

Does any part of Jmark take into account the video card??


64bit MC23 Jmark:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 4.185 seconds
    Single-threaded floating point math... 2.869 seconds
    Multi-threaded integer math... 2.345 seconds
    Multi-threaded mixed math... 1.500 seconds
Score: 1743

Running 'Image' benchmark...
    Image creation / destruction... 0.186 seconds
    Flood filling... 0.212 seconds
    Direct copying... 0.326 seconds
    Small renders... 0.965 seconds
    Bilinear rendering... 0.741 seconds
    Bicubic rendering... 0.423 seconds
Score: 7712

Running 'Database' benchmark...
    Create database... 0.133 seconds
    Populate database... 0.706 seconds
    Save database... 0.093 seconds
    Reload database... 0.032 seconds
    Search database... 1.011 seconds
    Sort database... 0.616 seconds
    Group database... 0.457 seconds
Score: 7052

JRMark (version 23.0.59): 5502

[Hendrik]

Does any part of Jmark take into account the video card??

No, its all CPU.

[tzr916]

Hmm, I wonder why not? Seems like it could be useful to add GPU benchmark into JRmark.

[Manfred]

64bit Version:

i7-4770T; MSI Z87M Gaming; 8 GB CL9 1600 MHz RAM; Samsung Evo SSD; MSI 1070 Gaming X:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 5,875 seconds
    Single-threaded floating point math... 3,648 seconds
    Multi-threaded integer math... 1,995 seconds
    Multi-threaded mixed math... 1,289 seconds
Score: 1484

Running 'Image' benchmark...
    Image creation / destruction... 0,270 seconds
    Flood filling... 0,483 seconds
    Direct copying... 0,503 seconds
    Small renders... 1,482 seconds
    Bilinear rendering... 1,162 seconds
    Bicubic rendering... 0,682 seconds
Score: 4802

Running 'Database' benchmark...
    Create database... 0,200 seconds
    Populate database... 1,388 seconds
    Save database... 0,396 seconds
    Reload database... 0,102 seconds
    Search database... 1,582 seconds
    Sort database... 1,029 seconds
    Group database... 0,706 seconds
Score: 3979

JRMark (version 23.0.62): 3422
-> Slower than old version 4207!

Old Results (32-bit Version)
4435  HTPC:  i3-6100; Asus B150I Pro Gaming; 8GB CL14 2133 MHz RAM; intel m.2 ssd; EVGA 960 SSC
4207  PC:      i7-4770T; MSI Z87M Gaming; 8 GB CL9 1600 MHz RAM; Samsung Evo SSD; MSI 1070 Gaming X:
1613: OldPC: Core2 6600, Nvidia nForce, 6 GB RAM 800 MHz:

[kensn]

My new build
I-7 3700
Asus p8z77-I
Adata s510 120g ssd
560TI

Stock speeds
JRiver fresh install with library backup..

Running 'Math' benchmark...
    Single-threaded integer math... 3.212 seconds
    Single-threaded floating point math... 2.132 seconds
    Multi-threaded integer math... 1.004 seconds
    Multi-threaded mixed math... 0.702 seconds
Score: 2695

Running 'Image' benchmark...
    Image creation / destruction... 0.704 seconds
    Flood filling... 0.357 seconds
    Direct copying... 0.671 seconds
    Small renders... 1.017 seconds
    Bilinear rendering... 0.703 seconds
    Bicubic rendering... 0.598 seconds
Score: 5432

Running 'Database' benchmark...
    Create database... 0.311 seconds
    Populate database... 1.070 seconds
    Save database... 0.118 seconds
    Reload database... 0.033 seconds
    Search database... 0.858 seconds
    Sort database... 0.845 seconds
    Group database... 0.622 seconds
Score: 5575

JRMark (version 17.0.182): 4568

64 bit today:

Running 'Math' benchmark...
    Single-threaded integer math... 3.225 seconds
    Single-threaded floating point math... 2.137 seconds
    Multi-threaded integer math... 0.976 seconds
    Multi-threaded mixed math... 0.643 seconds
Score: 2721

Running 'Image' benchmark...
    Image creation / destruction... 0.113 seconds
    Flood filling... 0.353 seconds
    Direct copying... 0.514 seconds
    Small renders... 0.981 seconds
    Bilinear rendering... 0.703 seconds
    Bicubic rendering... 0.429 seconds
Score: 7112

Running 'Database' benchmark...
    Create database... 0.157 seconds
    Populate database... 0.860 seconds
    Save database... 0.172 seconds
    Reload database... 0.074 seconds
    Search database... 1.125 seconds
    Sort database... 0.982 seconds
    Group database... 0.636 seconds
Score: 5366

JRMark (version 23.0.64 x64): 5067

Same PC... Seems like a nice bump ...

[JimH]

64bit Version:

i7-4770T; MSI Z87M Gaming; 8 GB CL9 1600 MHz RAM; Samsung Evo SSD; MSI 1070 Gaming X:
Score: 3979

JRMark (version 23.0.62): 3422
-> Slower than old version 4207!

Something isn't right.  Try rebooting.

[tyler69]

Apple MacBook10,1, Intel Core i7, 1,4 GHz, 16GB RAM, macOS 10.13

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 4.739 seconds
    Single-threaded floating point math... 3.267 seconds
    Multi-threaded integer math... 2.430 seconds
    Multi-threaded mixed math... 1.726 seconds
Score: 1562

Running 'Image' benchmark...
    Image creation / destruction... 1.678 seconds
    Flood filling... 0.305 seconds
    Direct copying... 1.420 seconds
    Small renders... 1.181 seconds
    Bilinear rendering... 1.436 seconds
    Bicubic rendering... 0.795 seconds
Score: 3228

Running 'Database' benchmark...
    Create database... 0.311 seconds
    Populate database... 1.420 seconds
    Save database... 0.170 seconds
    Reload database... 0.069 seconds
    Search database... 1.079 seconds
    Sort database... 0.887 seconds
    Group database... 0.760 seconds
Score: 4580

JRMark (version 23.0.65 x64): 3123

[jmone]

Stock : i7-8700K   

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3.224 seconds
    Single-threaded floating point math... 2.230 seconds
    Multi-threaded integer math... 0.659 seconds
    Multi-threaded mixed math... 0.469 seconds
Score: 2887

Running 'Image' benchmark...
    Image creation / destruction... 0.141 seconds
    Flood filling... 0.227 seconds
    Direct copying... 0.349 seconds
    Small renders... 0.738 seconds
    Bilinear rendering... 0.581 seconds
    Bicubic rendering... 0.323 seconds
Score: 9325

Running 'Database' benchmark...
    Create database... 0.106 seconds
    Populate database... 0.761 seconds
    Save database... 0.246 seconds
    Reload database... 0.076 seconds
    Search database... 0.802 seconds
    Sort database... 0.617 seconds
    Group database... 0.440 seconds
Score: 7054

JRMark (version 23.0.76 x64): 6422

[jmone]

i7-8700K using the ASUS (Idiot) Once Click Overcloker @ 5GHZ -     

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 2.955 seconds
    Single-threaded floating point math... 2.041 seconds
    Multi-threaded integer math... 0.590 seconds
    Multi-threaded mixed math... 0.455 seconds
Score: 3145

Running 'Image' benchmark...
    Image creation / destruction... 0.126 seconds
    Flood filling... 0.169 seconds
    Direct copying... 0.301 seconds
    Small renders... 0.677 seconds
    Bilinear rendering... 0.537 seconds
    Bicubic rendering... 0.295 seconds
Score: 10454

Running 'Database' benchmark...
    Create database... 0.098 seconds
    Populate database... 0.691 seconds
    Save database... 0.224 seconds
    Reload database... 0.058 seconds
    Search database... 0.790 seconds
    Sort database... 0.571 seconds
    Group database... 0.390 seconds
Score: 7614

JRMark (version 23.0.76 x64): 7071

[JimH]

Smoking...

[tyler69]

Would it make sense to offer a list or scale on what number is needed for what function? I remember seeing some comments like "XXXX should be enough for that" or "XXXX is pretty low for that" but an official categorization or scale vs. functions is what I mean.

[JimH]

It's not that simple.  What is "good" one year may only be "fair" a couple years later.

The lowest numbers here are probably still fine for ordinary audio, but DSD would require an above average machine.

The middle numbers are fine for ordinary video, and probably for ROHQ, but tweaking madVR might require higher values.  jmone just posted his advice on this subject.

The graphics card can also be a factor.

So can data problems related to drives that aren't internally connected.

Then there are antivirus programs.  They can cause a lot of performance related problems.

If I were to take a wild guess at it, I'd say a machine with a JRMark above 4000 is probably always going to work well.  Many machines below that may be OK.  Anything below 2000 may struggle.  Some will work.

[tzr916]

 Since JRmark does not benchmark the GPU, how do JRmark numbers have anything to do with ROHQ madVR? MadVR is pretty much all GPU, right?

[The Big Labinski]

Server - database and files

Benchmark 'Mathematik' wird ausgeführt...
    Single-Thread Ganzzahlmathematik... 3,761 Sekunden
    Single-Thread Gleitkommamathematik... 2,546 Sekunden
    Multi-Thread Ganzzahlmathematik... 2,262 Sekunden
    Multi-Thread gemischte Mathematik... 1,485 Sekunden
Score: 1890

Benchmark 'Bild' wird ausgeführt...
    Bild-Erstellung / -abbau... 0,272 Sekunden
    Flood filling... 0,387 Sekunden
    Direktes Kopieren... 0,468 Sekunden
    Small rendering... 1,247 Sekunden
    Bilinear rendering... 0,895 Sekunden
    Bicubic rendering... 0,530 Sekunden
Score: 5792

Benchmark 'Datenbank' wird ausgeführt...
    Datenbank erstellen... 0,311 Sekunden
    Datenbank befüllen... 1,321 Sekunden
    Datenbank speichern... 0,378 Sekunden
    Datenbank neu laden... 0,100 Sekunden
    Datenbank durchsuchen... 1,022 Sekunden
    Datenbank sortieren... 1,047 Sekunden
    Datenbank gruppieren... 0,688 Sekunden
Score: 4416

JRMark (version 23.0.77): 4033


Local PC - network access to database

Benchmark 'Mathematik' wird ausgeführt...
    Single-Thread Ganzzahlmathematik... 3,549 Sekunden
    Single-Thread Gleitkommamathematik... 2,179 Sekunden
    Multi-Thread Ganzzahlmathematik... 1,184 Sekunden
    Multi-Thread gemischte Mathematik... 0,716 Sekunden
Score: 2491

Benchmark 'Bild' wird ausgeführt...
    Bild-Erstellung / -abbau... 0,197 Sekunden
    Flood filling... 0,359 Sekunden
    Direktes Kopieren... 0,517 Sekunden
    Small rendering... 0,991 Sekunden
    Bilinear rendering... 0,775 Sekunden
    Bicubic rendering... 0,431 Sekunden
Score: 6729

Benchmark 'Datenbank' wird ausgeführt...
    Datenbank erstellen... 0,209 Sekunden
    Datenbank befüllen... 0,825 Sekunden
    Datenbank speichern... 0,164 Sekunden
    Datenbank neu laden... 0,075 Sekunden
    Datenbank durchsuchen... 0,841 Sekunden
    Datenbank sortieren... 0,787 Sekunden
    Datenbank gruppieren... 0,557 Sekunden
Score: 6216

JRMark (version 23.0.77): 5145

[haylo75]

I just put together a gaming machine for my nephew, and thought I'd run the benchmarks since it's the first machine I've put together in awhile.  Benchmarks respond well to overclocking CPU & RAM.


NO overclock
CPU: Core i5-7600k @ 3.8Ghz
RAM: Corsair Vengeance 2x8GB DDR4 @ 2.13Ghz
OS: Windows 10 x64
MC: 23.0.87 x64

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3.857 seconds
    Single-threaded floating point math... 2.640 seconds
    Multi-threaded integer math... 2.010 seconds
    Multi-threaded mixed math... 1.337 seconds
Score: 1930

Running 'Image' benchmark...
    Image creation / destruction... 0.192 seconds
    Flood filling... 0.241 seconds
    Direct copying... 0.387 seconds
    Small renders... 0.945 seconds
    Bilinear rendering... 0.708 seconds
    Bicubic rendering... 0.398 seconds
Score: 7661

Running 'Database' benchmark...
    Create database... 0.149 seconds
    Populate database... 0.895 seconds
    Save database... 0.252 seconds
    Reload database... 0.043 seconds
    Search database... 0.827 seconds
    Sort database... 0.741 seconds
    Group database... 0.464 seconds
Score: 6378

JRMark (version 23.0.87): 5323

-----


Overclocked
CPU: Core i5-7600k @ 4.5Ghz
RAM: Corsair Vengeance 2x8GB DDR4 @ 3.0Ghz
OS: Windows 10 x64
MC: 23.0.87 x64

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3.386 seconds
    Single-threaded floating point math... 2.329 seconds
    Multi-threaded integer math... 1.946 seconds
    Multi-threaded mixed math... 1.293 seconds
Score: 2122

Running 'Image' benchmark...
    Image creation / destruction... 0.174 seconds
    Flood filling... 0.179 seconds
    Direct copying... 0.304 seconds
    Small renders... 0.833 seconds
    Bilinear rendering... 0.628 seconds
    Bicubic rendering... 0.353 seconds
Score: 8904

Running 'Database' benchmark...
    Create database... 0.134 seconds
    Populate database... 0.787 seconds
    Save database... 0.207 seconds
    Reload database... 0.040 seconds
    Search database... 0.723 seconds
    Sort database... 0.650 seconds
    Group database... 0.434 seconds
Score: 7228

JRMark (version 23.0.87): 6084

[Awesome Donkey]

Here's my new 8700k at stock speeds. I'll overclock this CPU to 5.0 GHz in the next few days and then I'll post another benchmark.

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3.573 seconds
    Single-threaded floating point math... 2.462 seconds
    Multi-threaded integer math... 0.807 seconds
    Multi-threaded mixed math... 0.546 seconds
Score: 2572

Running 'Image' benchmark...
    Image creation / destruction... 0.344 seconds
    Flood filling... 0.198 seconds
    Direct copying... 0.369 seconds
    Small renders... 0.807 seconds
    Bilinear rendering... 0.691 seconds
    Bicubic rendering... 0.384 seconds
Score: 7876

Running 'Database' benchmark...
    Create database... 0.112 seconds
    Populate database... 0.815 seconds
    Save database... 0.262 seconds
    Reload database... 0.063 seconds
    Search database... 1.025 seconds
    Sort database... 0.523 seconds
    Group database... 0.411 seconds
Score: 6694

JRMark (version 23.0.98 x64): 5714

[Awesome Donkey]

Overclocked 8700k to 5.0 GHz. I guess the Meltdown/Spectre mitigations did cause a slight performance penalty compared to jmone's earlier scores, but I can live with that.

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3.064 seconds
    Single-threaded floating point math... 2.121 seconds
    Multi-threaded integer math... 0.585 seconds
    Multi-threaded mixed math... 0.434 seconds
Score: 3063

Running 'Image' benchmark...
    Image creation / destruction... 0.296 seconds
    Flood filling... 0.169 seconds
    Direct copying... 0.316 seconds
    Small renders... 0.700 seconds
    Bilinear rendering... 0.573 seconds
    Bicubic rendering... 0.327 seconds
Score: 9242

Running 'Database' benchmark...
    Create database... 0.100 seconds
    Populate database... 0.558 seconds
    Save database... 0.104 seconds
    Reload database... 0.047 seconds
    Search database... 0.906 seconds
    Sort database... 0.453 seconds
    Group database... 0.391 seconds
Score: 8402

JRMark (version 23.0.100 x64): 6902

[FireTech]

i5 6600K overclocked to 4.2 GHz

Running 'Math' benchmark...
    Single-threaded integer math... 3,667 seconds
    Single-threaded floating point math... 2,524 seconds
    Multi-threaded integer math... 2,041 seconds
    Multi-threaded mixed math... 1,358 seconds
Score: 1981

Running 'Image' benchmark...
    Image creation / destruction... 0,263 seconds
    Flood filling... 0,317 seconds
    Direct copying... 0,426 seconds
    Small renders... 0,909 seconds
    Bilinear rendering... 0,689 seconds
    Bicubic rendering... 0,391 seconds
Score: 7346

Running 'Database' benchmark...
    Create database... 0,147 seconds
    Populate database... 0,893 seconds
    Save database... 0,231 seconds
    Reload database... 0,043 seconds
    Search database... 0,834 seconds
    Sort database... 0,725 seconds
    Group database... 0,506 seconds
Score: 6364

JRMark (version 23.0.102): 5230

[haylo75]

I just built out one of the new Intel 8th generation NUCs, NUC8i7HVK.  Considering the form factor, I'm pleased with the performance.  Specs:

RAM: Kington HyperX Impact 32GB 3200MHz DDR4 CL20 SODIMM HX432S20IBK2/32
SSD for Windows: Samsung 960 PRO Series - 512GB PCIe NVMe - M.2 Internal SSD (MZ-V6P512BW)
SSD for Storage: Samsung 860 EVO 2TB M.2 SATA Internal SSD (MZ-N6E2T0BW)

Media Center is 64-bit.

This is on stock BIOS settings and High Performance profile in Windows 10.

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3.690 seconds
    Single-threaded floating point math... 2.540 seconds
    Multi-threaded integer math... 1.381 seconds
    Multi-threaded mixed math... 0.854 seconds
Score: 2245

Running 'Image' benchmark...
    Image creation / destruction... 0.340 seconds
    Flood filling... 0.163 seconds
    Direct copying... 0.286 seconds
    Small renders... 0.844 seconds
    Bilinear rendering... 0.727 seconds
    Bicubic rendering... 0.418 seconds
Score: 7919

Running 'Database' benchmark...
    Create database... 0.124 seconds
    Populate database... 0.816 seconds
    Save database... 0.249 seconds
    Reload database... 0.057 seconds
    Search database... 1.029 seconds
    Sort database... 0.908 seconds
    Group database... 0.695 seconds
Score: 5544

JRMark (version 24.0.27 64 bit): 5236

[JimH]

Nice score for a very small computer.

[PPC71454]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3.488 seconds
    Single-threaded floating point math... 2.406 seconds
    Multi-threaded integer math... 0.820 seconds
    Multi-threaded mixed math... 0.556 seconds
Score: 2614

Running 'Image' benchmark...
    Image creation / destruction... 0.435 seconds
    Flood filling... 0.211 seconds
    Direct copying... 0.401 seconds
    Small renders... 0.798 seconds
    Bilinear rendering... 0.416 seconds
    Bicubic rendering... 0.246 seconds
Score: 8773

Running 'Database' benchmark...
    Create database... 0.122 seconds
    Populate database... 0.764 seconds
    Save database... 0.242 seconds
    Reload database... 0.075 seconds
    Search database... 0.951 seconds
    Sort database... 0.519 seconds
    Group database... 0.418 seconds
Score: 6955

JRMark (version 24.0.45 64 bit): 6114

[Matt]

Built a new i9 9900k system.

Here's the bench:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 2.931 seconds
    Single-threaded floating point math... 2.024 seconds
    Multi-threaded integer math... 0.373 seconds
    Multi-threaded mixed math... 0.255 seconds
Score: 3404

Running 'Image' benchmark...
    Image creation / destruction... 0.475 seconds
    Flood filling... 0.199 seconds
    Direct copying... 0.375 seconds
    Small renders... 0.675 seconds
    Bilinear rendering... 0.260 seconds
    Bicubic rendering... 0.159 seconds
Score: 10267

Running 'Database' benchmark...
    Create database... 0.097 seconds
    Populate database... 0.647 seconds
    Save database... 0.196 seconds
    Reload database... 0.044 seconds
    Search database... 0.755 seconds
    Sort database... 0.445 seconds
    Group database... 0.339 seconds
Score: 8522

JRMark (version 25.0.8 64 bit): 7398

[kensn]

Huh.. Look at that! Let me know when you grow tired of that machine will ya?

[JimH]

I think that's a record.

[bob]

Raspberry Pi 4 running off of the SDCard.

=== Running Benchmarks (please do not interrupt) ===
Running 'Math' benchmark...
    Single-threaded integer math... 6.126 seconds
    Single-threaded floating point math... 4.671 seconds
    Multi-threaded integer math... 2.988 seconds
    Multi-threaded mixed math... 2.398 seconds
Score: 1174

Running 'Image' benchmark...
    Image creation / destruction... 2.542 seconds
    Flood filling... 2.573 seconds
    Direct copying... 2.582 seconds
    Small renders... 2.863 seconds
    Bilinear rendering... 11.854 seconds
    Bicubic rendering... 1.633 seconds
Score: 915

Running 'Database' benchmark...
    Create database... 1.590 seconds
    Populate database... 4.669 seconds
    Save database... 0.755 seconds
    Reload database... 0.215 seconds
    Search database... 3.970 seconds
    Sort database... 2.943 seconds
    Group database... 1.972 seconds
Score: 1334

JRMark (version 25.0.71): 1141

[NorthGeorgiaWX]

This is my HP Demon laptop.
i7-7700HQ @2.80GHz
---------------------------------------

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 4.367 seconds
    Single-threaded floating point math... 3.017 seconds
    Multi-threaded integer math... 1.506 seconds
    Multi-threaded mixed math... 0.933 seconds
Score: 1934

Running 'Image' benchmark...
    Image creation / destruction... 0.747 seconds
    Flood filling... 0.274 seconds
    Direct copying... 0.438 seconds
    Small renders... 1.010 seconds
    Bilinear rendering... 0.775 seconds
    Bicubic rendering... 0.452 seconds
Score: 5951

Running 'Database' benchmark...
    Create database... 0.146 seconds
    Populate database... 1.007 seconds
    Save database... 0.305 seconds
    Reload database... 0.077 seconds
    Search database... 1.242 seconds
    Sort database... 0.736 seconds
    Group database... 0.608 seconds
Score: 5218

JRMark (version 25.0.71 64 bit): 4368

[JimH]

Raspberry Pi 3, stock clock speeds, running on an SD card

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 9.217 seconds
    Single-threaded floating point math... 12.566 seconds
    Multi-threaded integer math... 6.066 seconds
    Multi-threaded mixed math... 6.730 seconds
Score: 549

Running 'Image' benchmark...
    Image creation / destruction... 1.965 seconds
    Flood filling... 4.748 seconds
    Direct copying... 4.398 seconds
    Small renders... 2.060 seconds
    Bilinear rendering... 3.675 seconds
    Bicubic rendering... 0.832 seconds
Score: 1245

Running 'Database' benchmark...
    Create database... 4.750 seconds
    Populate database... 9.453 seconds
    Save database... 1.073 seconds
    Reload database... 0.339 seconds
    Search database... 17.877 seconds
    Sort database... 7.876 seconds
    Group database... 4.534 seconds
Score: 468

JRMark (version 21.0.63): 754

The pi's are getting pretty close to some of the lower end intel hardware.  I expect the model 4 (unannounced and hypothetical) will be within striking distance of the Atom/Celerons.

mwillems was right again.  It just took a few years.

[mwillems]

mwillems was right again.  It just took a few years.

It's nice to be right eventually, although sometimes being right at the wrong time feels a lot like being wrong 

[wer]

It would really be helpful if JRMark added a section to incorporate graphics card performance relevant to MadVR, so as to give some idea of how well MadVR upscaling will perform.

Many view MadVR-assisted video playback as an essential part of MC, so does it make sense the benchmark totally ignores that?

[The Big Labinski]

Hi friends, got a new PC

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3,151 seconds
    Single-threaded floating point math... 2,178 seconds
    Multi-threaded integer math... 0,416 seconds
    Multi-threaded mixed math... 0,284 seconds
Score: 3151

Running 'Image' benchmark...
    Image creation / destruction... 0,222 seconds
    Flood filling... 0,242 seconds
    Direct copying... 0,424 seconds
    Small renders... 0,732 seconds
    Bilinear rendering... 0,300 seconds
    Bicubic rendering... 0,185 seconds
Score: 10450

Running 'Database' benchmark...
    Create database... 0,147 seconds
    Populate database... 0,550 seconds
    Save database... 0,080 seconds
    Reload database... 0,066 seconds
    Search database... 0,809 seconds
    Sort database... 0,458 seconds
    Group database... 0,406 seconds
Score: 8541

JRMark (version 25.0.96 64 bit): 7381

[JimH]

Nice.

[The Big Labinski]

Matt is 17 points better. I think I need a new PC - LOL.

[Manfred]

If you run the benchmark again you may get +18 more.  ?

[luvmich]

Just a fyi, Ryzen 9 3900x with 3200Mhz DDR4 cas 14,14,14,34
Not sure this benchmark works for 24 threads.  Only takes a few seconds to run the bench and my chip shows very little activity in task manager.  While running "Handbrake"  encoding a bluray my Jrmark is 4274.  Handbrake pegs my cpu to 100% on all 24 threads. 

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 1.549 seconds
    Single-threaded floating point math... 2.099 seconds
    Multi-threaded integer math... 0.351 seconds
    Multi-threaded mixed math... 0.285 seconds
Score: 4435

Running 'Image' benchmark...
    Image creation / destruction... 0.575 seconds
    Flood filling... 0.283 seconds
    Direct copying... 0.540 seconds
    Small renders... 0.727 seconds
    Bilinear rendering... 0.184 seconds
    Bicubic rendering... 0.138 seconds
Score: 8989

Running 'Database' benchmark...
    Create database... 0.121 seconds
    Populate database... 0.729 seconds
    Save database... 0.202 seconds
    Reload database... 0.126 seconds
    Search database... 0.928 seconds
    Sort database... 0.537 seconds
    Group database... 0.452 seconds
Score: 6947

JRMark (version 25.0.113 64 bit): 6790

[NorthGeorgiaWX]

Just tested my laptop. HP Omen i7.

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 4.453 seconds
    Single-threaded floating point math... 3.194 seconds
    Multi-threaded integer math... 1.497 seconds
    Multi-threaded mixed math... 0.944 seconds
Score: 1883

Running 'Image' benchmark...
    Image creation / destruction... 0.587 seconds
    Flood filling... 0.259 seconds
    Direct copying... 0.414 seconds
    Small renders... 1.029 seconds
    Bilinear rendering... 0.774 seconds
    Bicubic rendering... 0.460 seconds
Score: 6245

Running 'Database' benchmark...
    Create database... 0.158 seconds
    Populate database... 0.989 seconds
    Save database... 0.299 seconds
    Reload database... 0.078 seconds
    Search database... 1.239 seconds
    Sort database... 1.108 seconds
    Group database... 0.771 seconds
Score: 4634

JRMark (version 25.0.114 64 bit): 4254

[tzr916]

Why do JRMark scores drop every major version upgrade? On my laptop:
last version of MC24 JRMark ~ 2800.
last version of MC25 JRMark ~ 2400.
first version of MC26 JRMark ~ 2000.

Even more pronounced on my Desktop, lost over 900 points going from last version of MC 25 to first version of MC26!

[MrEWhite]

i9 9900KS @ 5.2 GHz, 16GB of DDR4 RAM at 4000 MHz, and a RTX 2080 Ti, beat this!
=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 2.927 seconds
    Single-threaded floating point math... 2.021 seconds
    Multi-threaded integer math... 0.382 seconds
    Multi-threaded mixed math... 0.255 seconds
Score: 3402

Running 'Image' benchmark...
    Image creation / destruction... 0.195 seconds
    Flood filling... 0.141 seconds
    Direct copying... 0.340 seconds
    Small renders... 0.679 seconds
    Bilinear rendering... 0.258 seconds
    Bicubic rendering... 0.161 seconds
Score: 12411

Running 'Database' benchmark...
    Create database... 0.159 seconds
    Populate database... 0.608 seconds
    Save database... 0.146 seconds
    Reload database... 0.033 seconds
    Search database... 0.825 seconds
    Sort database... 0.448 seconds
    Group database... 0.449 seconds
Score: 8058

JRMark (version 26.0.18 64 bit): 7957

[MrEWhite]

Something I noticed is that the database score on MC26 is lower than MC25.
Here's MC26:

Running 'Database' benchmark...
    Create database... 0.159 seconds
    Populate database... 0.608 seconds
    Save database... 0.146 seconds
    Reload database... 0.033 seconds
    Search database... 0.825 seconds
    Sort database... 0.448 seconds
    Group database... 0.449 seconds
Score: 8058

and here's MC25

Running 'Database' benchmark...
    Create database... 0.157 seconds
    Populate database... 0.593 seconds
    Save database... 0.154 seconds
    Reload database... 0.037 seconds
    Search database... 0.715 seconds
    Sort database... 0.418 seconds
    Group database... 0.407 seconds
Score: 8663

Weird.

Also, am I the first to crack 8000 score overall?

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 2.927 seconds
    Single-threaded floating point math... 2.021 seconds
    Multi-threaded integer math... 0.377 seconds
    Multi-threaded mixed math... 0.259 seconds
Score: 3403

Running 'Image' benchmark...
    Image creation / destruction... 0.220 seconds
    Flood filling... 0.141 seconds
    Direct copying... 0.354 seconds
    Small renders... 0.678 seconds
    Bilinear rendering... 0.253 seconds
    Bicubic rendering... 0.158 seconds
Score: 12205

Running 'Database' benchmark...
    Create database... 0.157 seconds
    Populate database... 0.593 seconds
    Save database... 0.154 seconds
    Reload database... 0.037 seconds
    Search database... 0.715 seconds
    Sort database... 0.418 seconds
    Group database... 0.407 seconds
Score: 8663

JRMark (version 25.0.115 64 bit): 8090

[ths61]

Aging Dell M6700 laptop.

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3.568 seconds
    Single-threaded floating point math... 2.370 seconds
    Multi-threaded integer math... 1.081 seconds
    Multi-threaded mixed math... 0.764 seconds
Score: 2441

Running 'Image' benchmark...
    Image creation / destruction... 0.184 seconds
    Flood filling... 0.394 seconds
    Direct copying... 0.438 seconds
    Small renders... 1.089 seconds
    Bilinear rendering... 0.763 seconds
    Bicubic rendering... 0.522 seconds
Score: 6491

Running 'Database' benchmark...
    Create database... 0.323 seconds
    Populate database... 1.036 seconds
    Save database... 0.293 seconds
    Reload database... 0.073 seconds
    Search database... 1.284 seconds
    Sort database... 0.938 seconds
    Group database... 0.592 seconds
Score: 4736

JRMark (version 26.0.12 64 bit): 4556

[tzr916]

Why do JRMark scores drop every major version upgrade? On my laptop:
last version of MC24 JRMark ~ 2800.
last version of MC25 JRMark ~ 2400.
first version of MC26 JRMark ~ 2000.

Even more pronounced on my Desktop, lost over 900 points going from last version of MC 25 to first version of MC26!

v26.0.20 Changed: Made the library creation portion of the benchmark faster again, also speeds up library creation in general.

My laptop is back up to 2895