PerformanceTest V10 Beta Release

[rofl]

scripting outputs seems the same as PT9 for me, I did not get the HIDEBASELINES command to work, maybe I'm using it incorrectly?

What I would like to see is the EXPORTCSV SINGLELINE formatted results for each iteration of the test.
EXPORTCSV SINGLELINE has the same results as PT9 - all hardware/OS details, the correct local time<as set on PC> and combined test run results into a single line.
REPORTSUMMARYCSV does have the individual run results with a number of extra lines, no HW/OS/Time details, and I could not get HIDEBASELINES to work, still have the baselines listed in the output file.

[Tim (PassMark)]

-It seems that I am still getting the PDF error when executing from script, or maybe it's intermittent, not enough testing completed to be conclusive.
"Unexpected error running PDF Render test. Error number -2147024894"

-Scripting multi-runs still stops and requires manually clicking OK at the end of a single run to continue

Could you please post the scripts you are using so we can test them here and see if we can reproduce the behaviour.

[HHH03]

I have been testing your Alpha Build V3 and my computer scores are much higher now vs build 1035. I'm curious as to how these Alpha Passmark scores will compare with Passmark V9. My CPU and NVME score is slightly lower in the Alpha build, but everything else is much higher... memory, 3D video, 2D video and the overall computer score. Unfortunately, I won't be able to use it much longer since my Passmark V9 key doesn't work with the Alpha version... only 7 days left.
I also tried to upload the results of my testing, but it doesn't show a real link to the result.
The biggest change is the 2D and 3D tests on my computer... the 3D test alone increased by over 5000 points vs Version 1035!

[rofl]

script attached

[Tim (PassMark)]

script attached

It doesn't look the script attached properly, could you please try again or post the script commands in a comment.

[rofl]

# Set the test preferences
SETDISK "C:"
SETDURATION 15
SETCOMPUTERNAME "MFG Model Ver"
SETCPUTESTPROCESSES 2
HIDEBASELINES
SET3DMAX 

# Run the test 5 times
LOOP 5
{

# Clear the results
CLEARRESULTS

# Run all tests 
RUN ALL


} REPORTSUMMARYCSV "C:\Results\sumresults.csv"


# Save a CSV to capture hardware, OS and time


EXPORTCSV "C:\Results\results.csv" SINGLELINE

[rofl]

Actually, getting the file format I'm looking for with this:

# Set the test preferences SETDISK "C:" SETDURATION 15 SETCOMPUTERNAME "MFG Model Ver" SETCPUTESTPROCESSES 2 HIDEBASELINES SET3DMAX   # Run the test 5 times LOOP 5 {  # Clear the results CLEARRESULTS  # Run all tests  RUN ALL  # Append to running CSV with results of single test run EXPORTCSV "C:\Results\results.csv" SINGLELINE }

Getting the PDF error so no 2D results and have to click OK to continue to next run.

[Tim (PassMark)]

HIDEBASELINES is used as part of the EXPORT commands, not as a standalone command (eg EXPORTHTML "<filename.html>" HIDEBASELINES).

Currently i don't believe it's valid with the REPORTSUMMARYCSV command so we'll look at adding it for the next release.

I believe we were able to reproduce the PDF error so we're looking into this now.

[rofl]

Also, I'm seeing some output errors. I'll validate this on some other system types that PT10 will hopefully pick up the disk info on but this is what I'm seeing for a few of the new columns:

...RAM,Drive Model,Drive Size,Drive Serial Number,Operating System...
..."16GB RAM","Unknown Disk","Unknown Disk",,"Unknown Disk",,,Windows 10...

For some reason there are extra commas in there after the Drive Size and Drive SN results.

[rofl]

Just follow-up and confirm.

I see the extra commas consistently when unknown disk (confirmed with 4 separate non-discovered systems), and no extra commas when disk is properly discovered (confirmed with 3 separate discovered systems):

..."32GB SK Hynix Unknown","PC SN730 NVMe WDC 512GB",476GB,1944B3800037,Windows 10...

no extra commas

[rofl]

Also also...

I never see the "Model" field populated in the CSV export. Is it supposed to return <COMPUTERNAME> as defined by the SETCOMPUTERNAME command?

[JohnG]

Getting error -2147024894 on PDF test

[Tim (PassMark)]

Also also...

I never see the "Model" field populated in the CSV export. Is it supposed to return <COMPUTERNAME> as defined by the SETCOMPUTERNAME command?

This field is actually only populated when saving a baseline (the "Computer model" field), we'll make a note to add a script command to set this field directly.

[Tim (PassMark)]

I see the extra commas consistently when unknown disk (confirmed with 4 separate non-discovered systems), and no extra commas when disk is properly discovered (confirmed with 3 separate discovered systemss

We've just been able to reproduce this so should have it fixed in the next build.

[rofl]

Also also also

1) I've noticed when writing to a NAS share a single PC seems to write and keep the file open. I was kinda hoping to have multiple PCs write to the same hosted file for a compiled CSV. - wish list item

2) Out-of-the-box using the script previously posted I'm getting very similar results on i7 and i9, more specifically with i7-1065G7 having as good or better results to i9-9880H. Is there a better way I can script testing max CPU throughput utilizing all cores at max clock for any type of CPU or is the 10th gen i7 really performing so well? I see you can specify number of core testing, but really the results I'm trying to compare are total throughput. Is the SETCPUTESTPROCESSES command used to set the number of cores and if so is "MAX" an option?