• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

"Last Man Standing" Post up your benching results from userbenchmark.com! (in house competition)

Joined
May 6, 2018
Messages
1,150 (0.53/day)
Location
Upstate NY
System Name Dual Socket HP z820 Workstation
Processor Twin Intel Xeon E5 2673 v2 OEM processors (thats a total of 16C/32T)
Motherboard HP Dual Socket Motherboard
Cooling Stock HP liquid cooling
Memory 64GB Registered ECC memory kit (octal channel memory on this rig)
Video Card(s) MSI RX 5700 XT Gaming X 8GB
Storage 2 x 512GB SSD in raid 0
Display(s) Acer 23" 75Hz Gaming monitors 1080P x2
Case Brushed Aluminium
Audio Device(s) Integrated (5.1)
Power Supply HP 1125W Stock PSU
Mouse gaming mouse
Keyboard Dell
Software Windows 10 Pro
I know, User benchmarking gets a bad rap, by some in the tech community. But if you can cut through all the BS you'll actually see that it has extremely extensive stats on various forms of relevant hardware. Their graphs and charts are second to none when it comes to mapping information. Making it very straightforward and easy to find the right CPU for your build. Best thing about it is it's driven purely by the end user so the data is not tainted in any way. What we are all really interested in is YOUR scores and if we get enough attention here I will start a leaderboard.

Basic Ground Rules:
Multiple rigs are allowed
Multiple submissions are allowed (we will use your highest rated result)
Laptops are NOT allowed
Slow systems are NOT allowed

*Overall WInner*
The overall winner of the competition will be determined by taking the average of both percentile rank and individual benchmark results. You can compete for highest percentile rank or individual benchmark rank as well, and respect will be paid, but the main determining factor in choosing our champion will be an average of the two.

So, considering the benchmarking software is free, readily available and fast to run, I highly recommend anyone reading this to please give it a chance. It's simple, just download the benchmark and let it run. Takes all of one minute. At the end it will display the results.... and compare them against other known hardware, making it very easy to compare and contrast and see your level of performance.

Here is a link to benchmarking software from userbenchmark.com:
User Benchmark Test

Please follow a general layout like this, can be adjusted if we think of something else or want to move things around.
**********************************************************************************************
-Provide a hyperlink to your benchmark result
-Provide a little background information on your rig or your specific overclock, if you'd like
-Any other relevant info you think we should know about your system

You may, or may not take snips of your results. If you have something exceptional and you want everyone to see it, by all means, post it with your result














*The overall winner of this competition will receive a fully functional AMD Phenom II 960T X4 CPU. This does NOT unlock into a hexacore, but still packs a punch and has an unlocked multi. If you want it.

*By no means are these rules set in stone, I just thought it would be cool to see how competitive we could make this. If you have any suggestions at all, I am all ears. So please, don't hesitate.

I think once we get our ducks in a row, this will make for some very interesting results, only time will tell!
 
Last edited:
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
Yikes. You loooove this benchmark, lol...MEH benchmark...PITA to capture images...liquid rules.... :lovetpu:

Note: this will take more than a minute if you have several drives and especially a USB stick plugged in (239 seconds without a USB drive attached - I do have 5 drives 3SSDs, 2HDDs). Take any USB sticks out first (we are unable to check off and not run that bench). I assume we'll take the fastest drive's results in cases like this?

Anyway, here are the 'snips' you asked for...this is 'stock' outside of the CPU clocks (NB Freq is stock, just multi and voltage). I also provided a link to the benchmark result which is a good idea to have as well... anyone can simply cut and paste results and win that phenom(inally old :p) CPU.

CPU:...…………….Intel i9 7960x @ 4.5GHz
Memory...……….32GB G.Skill Trident Z DDR4 3600
Motherboard....Asus ROG Rampage VI Extreme Omega
Cooling...………..3x120mm custom water

If I did this right, my average is 186.975. I added up the "This bench value" for CPU, Storage (best score for the HDD), GPU, and Memory.




EDIT: Also, I assume you are limiting participants to ambient cooled air and water and not including 'old' benchmarks (meaning they need to be from the date of this thread forward)? Hopefully nobody tested using userbench and LN2... or some won't get 100th percentile, lol!

EDIT2: I ran this a couple of times and the overall score varied by a couple of %... I've got a slower average score and a faster average score (sandbagged for later... :p).

EDIT3:
EDIT* Scoring and rank will be dictated by how well your system performs relative to similar hardware. Your score is essentially your ranking, going from 1-100th percentile. Because userbenchmark scores the benchmarks this way as well. AND your score will be determined based upon an average of your specific bench results (for example, my bench was 60.4 on my GPU. We would take every result like this, for each category, add them all up, then divide by 4. This will give us good information. And as I said, this is still open for debate in terms of how exactly we want to score this..

It should look clear when we start the leaderboard, you'll get the idea.

Ok, so I am confused. I would like to have seen a leaderboard with all of your results in it as an example... right now it's just a confusing passage of rules(suggestions?). At first you say it is how well your system performs relative to similar hardware... which means, it should go by the GREEN values where there is checkmark and says "Performing above expectations (82nd percentile)" - right? Yet, you said to use the 60.4 value which is "this bench" and appears to be against all submissions(? Do you know what that is for?). Can you clarify, please? Since you said 60.4 I used those ("this bench") values to give you my average.

It would make sense to me to use the green values as that is, as you said, compared to the same hardware and not against different setups. This way, the closest to 100 wins (but that begs the question, what if there is a tie???). You may want to figure this out sooner than later, lol.

I wish you can make a custom run or discard the HDD scores. They aren't tweakable and the results are essentially a testament to an having an emtpy or not full drive as opposed to tweaking to get the best out of the component. None of us have any control over that. Install device, install driver, enjoy. Whereas with CPU, RAM, and GPU, these are tweakable items. I vote to not use the storage scores at all...
 

Attachments

  • ub1.jpg
    ub1.jpg
    274.9 KB · Views: 361
  • ub2.jpg
    ub2.jpg
    159.7 KB · Views: 365
Last edited:
Joined
Nov 13, 2007
Messages
10,209 (1.71/day)
Location
Austin Texas
Processor 13700KF Undervolted @ 5.6/ 5.5, 4.8Ghz Ring 200W PL1
Motherboard MSI 690-I PRO
Cooling Thermalright Peerless Assassin 120 w/ Arctic P12 Fans
Memory 48 GB DDR5 7600 MHZ CL36
Video Card(s) RTX 4090 FE
Storage 2x 2TB WDC SN850, 1TB Samsung 960 prr
Display(s) Alienware 32" 4k 240hz OLED
Case SLIGER S620
Audio Device(s) Yes
Power Supply Corsair SF750
Mouse Xlite V2
Keyboard RoyalAxe
Software Windows 11
Benchmark Scores They're pretty good, nothing crazy.
Last edited:
Joined
May 4, 2011
Messages
633 (0.13/day)
System Name Smooth-Operator
Processor AMD Ryzen 7 3800x
Motherboard Asrock x570 Taichi
Cooling AMD Wraith Prism
Memory 2x16GB 3200MHz CL16@CL14 DDR4
Video Card(s) Sapphire Radeon RX 580 8GB NITRO+
Storage 2x4TB WD HGST 7K6 7200RPM 256MB
Display(s) Samsung S24E370DL 24" IPS Freesync 75Hz
Case Fractal Design Focus G Window Blue
Audio Device(s) Creative X-Fi Titanium PCIe x1
Power Supply Corsair HX850 80+ Platinum
Mouse Gigabyte Aorus M3
Keyboard Zalman ZM-K300M
Software Windows 10 x64 Enterprise/Ubuntu Budgie amd64
Joined
May 6, 2018
Messages
1,150 (0.53/day)
Location
Upstate NY
System Name Dual Socket HP z820 Workstation
Processor Twin Intel Xeon E5 2673 v2 OEM processors (thats a total of 16C/32T)
Motherboard HP Dual Socket Motherboard
Cooling Stock HP liquid cooling
Memory 64GB Registered ECC memory kit (octal channel memory on this rig)
Video Card(s) MSI RX 5700 XT Gaming X 8GB
Storage 2 x 512GB SSD in raid 0
Display(s) Acer 23" 75Hz Gaming monitors 1080P x2
Case Brushed Aluminium
Audio Device(s) Integrated (5.1)
Power Supply HP 1125W Stock PSU
Mouse gaming mouse
Keyboard Dell
Software Windows 10 Pro
Yikes. You loooove this benchmark, lol...MEH benchmark...PITA to capture images...liquid rules.... :lovetpu:

Note: this will take more than a minute if you have several drives and especially a USB stick plugged in (239 seconds without a USB drive attached - I do have 5 drives 3SSDs, 2HDDs). Take any USB sticks out first (we are unable to check off and not run that bench). I assume we'll take the fastest drive's results in cases like this?

Anyway, here are the 'snips' you asked for...this is 'stock' outside of the CPU clocks (NB Freq is stock, just multi and voltage). I also provided a link to the benchmark result which is a good idea to have as well... anyone can simply cut and paste results and win that phenom(inally old :p) CPU.

CPU:...…………….Intel i9 7960x @ 4.5GHz
Memory...……….32GB G.Skill Trident Z DDR4 3600
Motherboard....Asus ROG Rampage VI Extreme Omega
Cooling...………..3x120mm custom water

If I did this right, my average is 186.975. I added up the "This bench value" for CPU, Storage (best score for the HDD), GPU, and Memory.




EDIT: Also, I assume you are limiting participants to ambient cooled air and water and not including 'old' benchmarks (meaning they need to be from the date of this thread forward)? Hopefully nobody tested using userbench and LN2... or some won't get 100th percentile, lol!

EDIT2: I ran this a couple of times and the overall score varied by a couple of %... I've got a slower average score and a faster average score (sandbagged for later... :p).

EDIT3:


Ok, so I am confused. I would like to have seen a leaderboard with all of your results in it as an example... right now it's just a confusing passage of rules(suggestions?). At first you say it is how well your system performs relative to similar hardware... which means, it should go by the GREEN values where there is checkmark and says "Performing above expectations (82nd percentile)" - right? Yet, you said to use the 60.4 value which is "this bench" and appears to be against all submissions(? Do you know what that is for?). Can you clarify, please? Since you said 60.4 I used those ("this bench") values to give you my average.

It would make sense to me to use the green values as that is, as you said, compared to the same hardware and not against different setups. This way, the closest to 100 wins (but that begs the question, what if there is a tie???). You may want to figure this out sooner than later, lol.

I wish you can make a custom run or discard the HDD scores. They aren't tweakable and the results are essentially a testament to an having an emtpy or not full drive as opposed to tweaking to get the best out of the component. None of us have any control over that. Install device, install driver, enjoy. Whereas with CPU, RAM, and GPU, these are tweakable items. I vote to not use the storage scores at all...

First and foremost, this benchmark yields a number of data points. I wanted to first get the word out to gauge interest and then come to a consensus as a group on how we would score the results. It will take some time and effort to come up with a leaderboard template, so I held off on creating it until we have all our ducks in a row. Plus, I didn't want to waste time by creating the leaderboard first only to find out later there was no interest. But it looks like we have some interest here so let's fine tune our competition.

*YOU WILL BE COMPETING AGAINST SIMILAR HARDWARE* SO THIS IS NOT A DIRECT ASSESSMENT OF YOUR COMPUTER'S PERFORMANCE RELATIVE TO ANOTHER FORUM MEMBERS COMPUTERS. IT'S A DIRECT ASSESSMENT OF YOUR COMPUTER'S PERFORMANCE RELATIVE TO ASSESSMENTS DONE BY USERBENCHMARK OF SIMILAR HARDWARE.

THE NEXT QUESTION TO ADDRESS IS HOW THEN WILL WE BE COMPETITING AGAINST EACH OTHER?


First, I did bring up two data sets that will potentially serve as "rubrics" or methods of scoring for this competition... The first being your PC performance level result as determined by the percentile rank, relative to similar hardware. For example, if we look at my result, my score/rank in this case would be 99. The closer to 100 you get, the better. If we have a tie, we will go to the second data set, which I will outline below.

First method of Scoring by percentile:
1582830290561.png


Second method (the assessment of your individual benchmark result) of scoring will be the average % of "This bench" results for your CPU, Graphics Card, SSD and Memory (so CPU + Graphics Card + SSD + Memory / 4). This data set is a key indicator of technical skill and overclock ability, essentially, this will assess or rate your technical tuning skills and your ability to maximize the performance potential of your hardware configuration. The higher your average, the better.

For example, in my case, it would be 113 (CPU) + 60.4 (Graphics Card) + 417 (SSD) + 131 (Memory) / 4 = 180.35

So, how will we determine the winner if we have two data sets?

The overall winner of the competition will be determined by adding your percentage result and your individual benchmark result and finding the average. In my case, that would be 99 (percentile score) + 180.35 (individual benchmark score) and then dividing by two. In my case, that would work out to 139.675. And this is the number that will drive the leaderboard, the higher the better.


Respect will be given for top ranking computers in both percentile score and individual score, but the average of these two is the key to success and that result will carry the most weight.

Sound okay to everyone? You guys think this approach will work out alright? I will frame up the leaderboard this afternoon. Expect something up by late evening.

By all means, let me know if there is anything else we need to do here to bring clarity to our results. Im all ears.

And yes, a link to your results is the best method to submit your data. Forget about snipping your results, just post the hyperlink

Since the results will have your system specs listed, there is no need to provide that data. Unless you want to point some details out or explain your overclock setup for us.
 
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
Ok, I understand better now, thank you for being clear.


That said, I get the first value is the average and the primary method for determining a winner (the average of the bold green scores, and what the little image is the average of those)...until there is a tie and the second method comes in. Here is where we diverge...

You say:
Second method (the assessment of your individual benchmark result) of scoring will be the average % of "This bench" results for your CPU, Graphics Card, SSD and Memory (so CPU + Graphics Card + SSD + Memory / 4). This data set is a key indicator of technical skill and overclock ability, essentially, this will assess or rate your technical tuning skills and your ability to maximize the performance potential of your hardware configuration. The higher your average, the better.
However, the second method (using "this bench" values) is how your hardware scores against all hardware tested for that section (drives/cpu/gpu/memory). It does NOT seem to be an indicator of technical skill and overclocking abilities/tuning/maximize, etc. The FIRST value does that as it is the ONLY comparison between LIKE hardware. The method is fine, but assuming I am reading this stuff right, you have it backwards on what results show. Where that first average value is based off the bold/green is what tells you how well you tweaked your hardware against similar hardware. As you said earlier, this isn't about how you did against other people's machines.......but your scoring, the second method appears to do just that.

In other words, in the image below I believe it means the following:
Blue arrow = average score for all Sammy 970 Pros against all drive submissions (inc. HDD/SSD etc).
Green arrow = how YOUR single Sammy 970 Pro did against all other 970 Pros only
Pink ameoba = how YOUR single sammy 970 Pro did against all drive submissions (inc. HDD/SSD etc).
img2.png


Is there a legend or something at the userbench site? I know it says x (horizontal) axis is score and y (vertical) is the number of submissions at a given result... but it doesn't clarify what I've pointed out above. Sorry if I am being particularly dense here, but just making sure you are actually getting what it is you want out of this thread and results. :)

EDIT: I'd also like to see drive scores dropped altogether. Again we can't tweak these... so why should anyone's scores be bolstered or held back in a contest specifically designed to compare "technical skill and overclocking abilities/tuning/maximize". Again, its install device, install driver, viola. There is no tweaking involved on a drive to yield better results. Look at what Acrtucas' drive below did to his overall average score...something is wrong there... it isn't because it wasn't tweaked, ya know? His saving grace in that run is the USB attached NVMe!! Do you see why this drive test stinks?

I also noticed those with mixed drives are likely to be penalized. In my case, I have a mix of 2x M.2 PCIe NVMe drives (OS and Games), a single SATA SSD (game backups), and two spinners for warm/cold storage. Scores like that end up hurting those with a mixed setup while those without HDDs or That also has nothing to do with maximizing performance or tweaking. I mean, users in this situation will have to disable all other drives so it only picks up the fastest I guess (though I can't do that anyway as it isn't my OS drive according to this test).

I say to drop drive scores..useless takes away from what you want to get out of these results. :)
 
Last edited:
Joined
Jul 14, 2006
Messages
2,405 (0.37/day)
Location
People's Republic of America
System Name It's just a computer
Processor i9-9900K Direct Die
Motherboard eVGA Z390 Dark
Cooling Dual D5T Vario, XSPC BayRes, Nemesis GTR560, NF-A14-iPPC3000PWM, NF-A14-iPPC2000, HK IV Pro Nickel
Memory G.Skill F4-4500C19D-16GTZKKE or G.Skill F4-3600C16D-16GTZ or G.Skill F4-4000C19D-32GTZSW
Video Card(s) eVGA RTX2080 FTW3 Ultra
Storage Samsung 960 EVO M.2
Display(s) LG 32GK650F
Case Thermaltake Xaser VI
Audio Device(s) Auzentech X-Meridian 7.1 2G/Z-5500
Power Supply Seasonic Prime PX-1300
Mouse Logitech
Keyboard Logitech
Software Win7 Ultimate x64 SP1
Last edited:
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
Seems the nvme drive is the problem.
Yep, said that above about your NVMe when I saw the result. That........and the bench itself can be wonky/inconsistent with some benchmarks. It's seem best to run multiple times and 'pop' the best score. When all four components align in a holy quad for the best score! :p
 
Joined
May 6, 2018
Messages
1,150 (0.53/day)
Location
Upstate NY
System Name Dual Socket HP z820 Workstation
Processor Twin Intel Xeon E5 2673 v2 OEM processors (thats a total of 16C/32T)
Motherboard HP Dual Socket Motherboard
Cooling Stock HP liquid cooling
Memory 64GB Registered ECC memory kit (octal channel memory on this rig)
Video Card(s) MSI RX 5700 XT Gaming X 8GB
Storage 2 x 512GB SSD in raid 0
Display(s) Acer 23" 75Hz Gaming monitors 1080P x2
Case Brushed Aluminium
Audio Device(s) Integrated (5.1)
Power Supply HP 1125W Stock PSU
Mouse gaming mouse
Keyboard Dell
Software Windows 10 Pro
Ok, I understand better now, thank you for being clear.


That said, I get the first value is the average and the primary method for determining a winner (the average of the bold green scores, and what the little image is the average of those)...until there is a tie and the second method comes in. Here is where we diverge...

You say:

However, the second method (using "this bench" values) is how your hardware scores against all hardware tested for that section (drives/cpu/gpu/memory). It does NOT seem to be an indicator of technical skill and overclocking abilities/tuning/maximize, etc. The FIRST value does that as it is the ONLY comparison between LIKE hardware. The method is fine, but assuming I am reading this stuff right, you have it backwards on what results show. Where that first average value is based off the bold/green is what tells you how well you tweaked your hardware against similar hardware. As you said earlier, this isn't about how you did against other people's machines.......but your scoring, the second method appears to do just that.

In other words, in the image below I believe it means the following:
Blue arrow = average score for all Sammy 970 Pros against all drive submissions (inc. HDD/SSD etc).
Green arrow = how YOUR single Sammy 970 Pro did against all other 970 Pros only
Pink ameoba = how YOUR single sammy 970 Pro did against all drive submissions (inc. HDD/SSD etc).
View attachment 146140

Is there a legend or something at the userbench site? I know it says x (horizontal) axis is score and y (vertical) is the number of submissions at a given result... but it doesn't clarify what I've pointed out above. Sorry if I am being particularly dense here, but just making sure you are actually getting what it is you want out of this thread and results. :)

EDIT: I'd also like to see drive scores dropped altogether. Again we can't tweak these... so why should anyone's scores be bolstered or held back in a contest specifically designed to compare "technical skill and overclocking abilities/tuning/maximize". Again, its install device, install driver, viola. There is no tweaking involved on a drive to yield better results. Look at what Acrtucas' drive below did to his overall average score...something is wrong there... it isn't because it wasn't tweaked, ya know? Drop drive scores man...useless and in fact takes away from what you want to get out of these results.

This is exactly what I wanted to have happen so we have a good competition with good results.

So the two data sets we are working with are:
1) Percentile Score - Determined by this number ONLY (highlighted in yellow) - we are NOT using individual percentile per piece of hardware, this is OVERALL percentile ONLY based on the system as a whole and shows up looking like this (highlighted in yellow) above your individual benchmark scores
1582845491443.png


2) Individual Bench Scores - Initially, I wanted this to be an average of your "Bench" scores but that's not going to tell us anything.

How about, we assess this result based on the difference between the "average bench (94%)" and your actual "This Bench score (113%)"
1582846422297.png


In this case, my result of 113% is 19 points above the average bench for this CPU - this increase over the "average bench" is what I mean when I talk about technical skill and overclocking abilities. I would then go through the remaining list of hardware and calculate the performance level of each part using the same method. These numbers would then be averaged to find the overall performance level of your system over what userbenchmark considers as "average".


Yep, said that above about your NVMe when I saw the result.
No, you need to have a fast SSD if you want to compete in the competition, points will be taken away because disk I/0 is a very important component and cannot be overlooked.
 
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
This is exactly what I wanted to have happen so we have a good competition with good results.

So the two data sets we are working with are:
1) Percentile Score - Determined by this number ONLY (highlighted in yellow) - we are NOT using individual percentile per piece of hardware, this is OVERALL percentile ONLY based on the system as a whole and shows up looking like this (highlighted in yellow) above your individual benchmark scores
View attachment 146145

2) Individual Bench Scores - Initially, I wanted this to be an average of your "Bench" scores but that's not going to tell us anything.

How about, we assess this result based on the difference between the "average bench (94%)" and your actual "This Bench score (113%)"
View attachment 146150

In this case, my result of 113% is 19 points above the average bench for this CPU - this increase over the "average bench" is what I mean when I talk about technical skill and overclocking abilities. I would then go through the remaining list of hardware and calculate the performance level of each part using the same method. These numbers would then be averaged to find the overall performance level of your system over what userbenchmark considers as "average".



No, you need to have a fast SSD if you want to compete in the competition, points will be taken away because disk I/0 is a very important component and cannot be overlooked.
1. The percentile is derived from the average of the individual scores, so you are in fact using each individual score (green/bold) there, note...
2. That's seems ok at first glance. But the value is still based on ALL storage submissions instead of against like hardware.

I agree fast storage is important... however this storage test has issues and you need to figure out a way around that...
A. Test variability... look at arcturus score in the same drive from one to the next. (Did I mention userbench is a bad test? :p)
B. You cant tweak storage. Doing well in storage depends on $$$$ for a single fast nvme and the ability to test using a fresh drive for best results. The antithesis of what you are trying to gauge here.
C. Keeping my #1 in mind, if you have a mix of fast and slow drives, you get penalized for the overall average by these individual results. Look at my link. Why should anyone get penalized because they have hdd for media and cold storage? My OS and game storage is nvme... yet spinners kill my score.
D. Arcturus' fastest drive seems like usb attached external storage... not tweakable and not a part of the system to make it faster.. why should that count, external storage, good or bad?

My only answer is to get rid of it or only use the fastest result. The problem with the latter is the fastest percentile score may not be the fastest "my bench" score...so storage is all wonky no? Seems easier to drop it imo...a lot of effort to maintain and seems to go against the spirit of your thread.
 
Last edited:
Joined
Jul 14, 2006
Messages
2,405 (0.37/day)
Location
People's Republic of America
System Name It's just a computer
Processor i9-9900K Direct Die
Motherboard eVGA Z390 Dark
Cooling Dual D5T Vario, XSPC BayRes, Nemesis GTR560, NF-A14-iPPC3000PWM, NF-A14-iPPC2000, HK IV Pro Nickel
Memory G.Skill F4-4500C19D-16GTZKKE or G.Skill F4-3600C16D-16GTZ or G.Skill F4-4000C19D-32GTZSW
Video Card(s) eVGA RTX2080 FTW3 Ultra
Storage Samsung 960 EVO M.2
Display(s) LG 32GK650F
Case Thermaltake Xaser VI
Audio Device(s) Auzentech X-Meridian 7.1 2G/Z-5500
Power Supply Seasonic Prime PX-1300
Mouse Logitech
Keyboard Logitech
Software Win7 Ultimate x64 SP1
The more I read and re-read the scoring rules, the more confused I am.

Basically, the score is determined by averaging the individual component scores and then taking the average of that score and the overall percentile score to get the final score, correct?

And again, can someone answer my question about 'sustained writes'?
 
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
The more I read and re-read the scoring rules, the more confused I am.

Basically, the score is determined by averaging the individual component scores and then taking the average of that score and the overall percentile score to get the final score, correct?

And again, can someone answer my question about 'sustained writes'?
I get the waiting for sustained writes, yes... a couple of times.. I think it's my spinners.
 
Joined
Jul 14, 2006
Messages
2,405 (0.37/day)
Location
People's Republic of America
System Name It's just a computer
Processor i9-9900K Direct Die
Motherboard eVGA Z390 Dark
Cooling Dual D5T Vario, XSPC BayRes, Nemesis GTR560, NF-A14-iPPC3000PWM, NF-A14-iPPC2000, HK IV Pro Nickel
Memory G.Skill F4-4500C19D-16GTZKKE or G.Skill F4-3600C16D-16GTZ or G.Skill F4-4000C19D-32GTZSW
Video Card(s) eVGA RTX2080 FTW3 Ultra
Storage Samsung 960 EVO M.2
Display(s) LG 32GK650F
Case Thermaltake Xaser VI
Audio Device(s) Auzentech X-Meridian 7.1 2G/Z-5500
Power Supply Seasonic Prime PX-1300
Mouse Logitech
Keyboard Logitech
Software Win7 Ultimate x64 SP1
Joined
Jul 14, 2006
Messages
2,405 (0.37/day)
Location
People's Republic of America
System Name It's just a computer
Processor i9-9900K Direct Die
Motherboard eVGA Z390 Dark
Cooling Dual D5T Vario, XSPC BayRes, Nemesis GTR560, NF-A14-iPPC3000PWM, NF-A14-iPPC2000, HK IV Pro Nickel
Memory G.Skill F4-4500C19D-16GTZKKE or G.Skill F4-3600C16D-16GTZ or G.Skill F4-4000C19D-32GTZSW
Video Card(s) eVGA RTX2080 FTW3 Ultra
Storage Samsung 960 EVO M.2
Display(s) LG 32GK650F
Case Thermaltake Xaser VI
Audio Device(s) Auzentech X-Meridian 7.1 2G/Z-5500
Power Supply Seasonic Prime PX-1300
Mouse Logitech
Keyboard Logitech
Software Win7 Ultimate x64 SP1
Joined
May 6, 2018
Messages
1,150 (0.53/day)
Location
Upstate NY
System Name Dual Socket HP z820 Workstation
Processor Twin Intel Xeon E5 2673 v2 OEM processors (thats a total of 16C/32T)
Motherboard HP Dual Socket Motherboard
Cooling Stock HP liquid cooling
Memory 64GB Registered ECC memory kit (octal channel memory on this rig)
Video Card(s) MSI RX 5700 XT Gaming X 8GB
Storage 2 x 512GB SSD in raid 0
Display(s) Acer 23" 75Hz Gaming monitors 1080P x2
Case Brushed Aluminium
Audio Device(s) Integrated (5.1)
Power Supply HP 1125W Stock PSU
Mouse gaming mouse
Keyboard Dell
Software Windows 10 Pro
Okay Guys. Im going to start on the leaderboard tonight --- Things will come together here.

What you need to know: Fine tune your system and run the benchmark.

You can post multiple runs, just be sure to tell me exactly which one you want me to use.

1. The percentile is derived from the average of the individual scores, so you are in fact using each individual score (green/bold) there, note...
2. That's seems ok at first glance. But the value is still based on ALL storage submissions instead of against like hardware.

I agree fast storage is important... however this storage test has issues and you need to figure out a way around that...
A. Test variability... look at arcturus score in the same drive from one to the next. (Did I mention userbench is a bad test? :p)
B. You cant tweak storage. Doing well in storage depends on $$$$ for a single fast nvme and the ability to test using a fresh drive for best results. The antithesis of what you are trying to gauge here.
C. Keeping my #1 in mind, if you have a mix of fast and slow drives, you get penalized for the overall average by these individual results. Look at my link. Why should anyone get penalized because they have hdd for media and cold storage? My OS and game storage is nvme... yet spinners kill my score.
D. Arcturus' fastest drive seems like usb attached external storage... not tweakable and not a part of the system to make it faster.. why should that count, external storage, good or bad?

My only answer is to get rid of it or only use the fastest result. The problem with the latter is the fastest percentile score may not be the fastest "my bench" score...so storage is all wonky no? Seems easier to drop it imo...a lot of effort to maintain and seems to go against the spirit of your thread.

I think you are reading too far into this earthdog. There is an expression that can be applied here and abbreviated KISS. That's what we need to focus on here. EVERYONE ELSE:

Get your system tuned and submit your best result. This will come together in the next few days.

Thanks!

Just a test run for our leaderboard... continuing to refine... I wish we could get more meaningful numbers but we have to work with what we've got, that's all there is too it.
So your "Performance AVG" is calculated by adding your score from CPU performance, Memory perf, HDD/SSD Perf, and graphics card perf and then dividing by 4. More specifically, your performance average is calculated from the difference between the "average bench (overall rating of this piece of specific hardware)" and "This bench" (your individual result). The higher your performance average, the higher your ranking on the leaderboard.

Can definitely be refined more, but at least it's a start...


1582942614580.png
 
Last edited:
Joined
Apr 15, 2013
Messages
1,261 (0.32/day)
System Name Some computer stuff
Processor Mostly Intel or AMD
Motherboard ATX or mATX
Cooling Bong Cooler
Memory DDR2-4
Video Card(s) A few
Storage Plenty Platters or SSDs or USBs
Display(s) Samsung 23"
Case 5 on the floor
Audio Device(s) There's one for my M7 Gene, Oh I have 3-4 PCI 5.1 ones.Sabrent! lol
Power Supply 750-1000W
Mouse cheap
Keyboard Used ps2 from garage sales
Software Yeah
Benchmark Scores http://hwbot.org/user/schmuckley/#Hardware_Library http://valid.canardpc.com/rbjpbg
^Not bad. What does user rank "potato" mean?

Ryzen 5 3600X @4.4-ish megglehertz. Almost a gaming machine.


UserBenchmarks: Game 72%, Desk 84%, Work 67%
CPU: AMD Ryzen 5 3600X - 99.8%
GPU: Nvidia GTX 1660 - 71.6%
SSD: Intel SSD Pro 1500 Series 180GB - 73.1%
SSD: Intel SSD Pro 1500 Series 180GB - 73.6%
RAM: G.SKILL F4-3000C15-8GVRB F4-3200C16-8GVGB F4-3000C15-8GVRB F4-3200C16-8GVGB 32GB - 90.3%
MBD: Asus ROG CROSSHAIR VII HERO
 
Last edited:
Joined
May 6, 2018
Messages
1,150 (0.53/day)
Location
Upstate NY
System Name Dual Socket HP z820 Workstation
Processor Twin Intel Xeon E5 2673 v2 OEM processors (thats a total of 16C/32T)
Motherboard HP Dual Socket Motherboard
Cooling Stock HP liquid cooling
Memory 64GB Registered ECC memory kit (octal channel memory on this rig)
Video Card(s) MSI RX 5700 XT Gaming X 8GB
Storage 2 x 512GB SSD in raid 0
Display(s) Acer 23" 75Hz Gaming monitors 1080P x2
Case Brushed Aluminium
Audio Device(s) Integrated (5.1)
Power Supply HP 1125W Stock PSU
Mouse gaming mouse
Keyboard Dell
Software Windows 10 Pro
n expression that can be applied here and abbreviated KISS. That's what we need to focus on here. EVERYONE ELSE:

Get your system tuned and submit your best result. This will come together in the next few days.
^Not bad. What does user rank "potato" mean?

Ryzen 5 3600X @4.4-ish megglehertz. Almost a gaming machine.


UserBenchmarks: Game 72%, Desk 84%, Work 67%
CPU: AMD Ryzen 5 3600X - 99.8%
GPU: Nvidia GTX 1660 - 71.6%
SSD: Intel SSD Pro 1500 Series 180GB - 73.1%
SSD: Intel SSD Pro 1500 Series 180GB - 73.6%
RAM: G.SKILL F4-3000C15-8GVRB F4-3200C16-8GVGB F4-3000C15-8GVRB F4-3200C16-8GVGB 32GB - 90.3%
MBD: Asus ROG CROSSHAIR VII HERO

User rank "potato"?

Why that's you, of course ;)
 
Joined
Jul 14, 2006
Messages
2,405 (0.37/day)
Location
People's Republic of America
System Name It's just a computer
Processor i9-9900K Direct Die
Motherboard eVGA Z390 Dark
Cooling Dual D5T Vario, XSPC BayRes, Nemesis GTR560, NF-A14-iPPC3000PWM, NF-A14-iPPC2000, HK IV Pro Nickel
Memory G.Skill F4-4500C19D-16GTZKKE or G.Skill F4-3600C16D-16GTZ or G.Skill F4-4000C19D-32GTZSW
Video Card(s) eVGA RTX2080 FTW3 Ultra
Storage Samsung 960 EVO M.2
Display(s) LG 32GK650F
Case Thermaltake Xaser VI
Audio Device(s) Auzentech X-Meridian 7.1 2G/Z-5500
Power Supply Seasonic Prime PX-1300
Mouse Logitech
Keyboard Logitech
Software Win7 Ultimate x64 SP1
Last edited:
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)

CPU @ 5500MHz RAM @ 4500 MHz 17-18-18-37.

Let me see if I have this correct.

The above benchmark would yield a score of 27.75?

Where does the 94th percentile come into play?

Also, I must agree with @EarthDog, the HDD/SSD might be an issue, since the drives are not 'overclockable'.
According to the new scoring method, it doesn't factor in at all? Not sure, it's still a bit convoluted for me too.

RE: the drives, the entire point/intent/spirit of this thread seems to have taken a turn once the OP understood what the values mean. So it isn't so much about tweaking really. Because of how the scoring system now appears to be setup, 25% of the result is based on a non-tweakable piece of hardware.

As far as sustained writes, I think that is a 60s test of sustained writes to the storage.
 
Last edited:
Joined
Apr 14, 2019
Messages
221 (0.12/day)
System Name Violet
Processor AMD Ryzen 5800X
Motherboard ASRock x570 Phantom Gaming X
Cooling Be quiet! Dark Rock Pro 4
Memory G.Skill Flare x 32GB 3400Mhz
Video Card(s) MSI 6900XT Gaming X Trio
Storage Western Digital WD Black SN750 1TB
Display(s) 3440x1440
Case Lian Li LANCOOL II MESH Performance
Power Supply Corsair RM850x
Mouse EVGA X15
Keyboard Corsair K95 RGB
Software Windows 10 64bit
userbenchmark.com......lol :wtf:

 
Last edited:
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
userbenchmark.com......lol :wtf:

From 25 mins in... that says it all. Storm-chaser, I implore you to have this on and listen to it.. if only from ~25 minutes in to the end.

"it's not that the website is now bad (because of its changes)... it was never good".

'should not be a resource for buying decisions...it's just not accurate'

"our audience isn't the audience that uses userbenchmarks.com" (inferring that enthusiasts are aware this isn't a good test but is for 'potatos')

I forgot GamersNexus also ripped into this site too.
https://www.reddit.com/r/buildapc/comments/chco8h
Best thing about it is it's driven purely by the end user so the data is not tainted in any way.
Knowing is half the battle, ehh? Otherwise it's just red and blue lasers (looks at my avatar)....

They tainted it. I told you this benchmark was shyte! I didn't know about the changes they made last year that made it worse. Respectfully, I agree with HW Unboxed, GN, and many others in the know. I'm out. GL to you, SC. :)
 
Last edited:
Joined
May 6, 2018
Messages
1,150 (0.53/day)
Location
Upstate NY
System Name Dual Socket HP z820 Workstation
Processor Twin Intel Xeon E5 2673 v2 OEM processors (thats a total of 16C/32T)
Motherboard HP Dual Socket Motherboard
Cooling Stock HP liquid cooling
Memory 64GB Registered ECC memory kit (octal channel memory on this rig)
Video Card(s) MSI RX 5700 XT Gaming X 8GB
Storage 2 x 512GB SSD in raid 0
Display(s) Acer 23" 75Hz Gaming monitors 1080P x2
Case Brushed Aluminium
Audio Device(s) Integrated (5.1)
Power Supply HP 1125W Stock PSU
Mouse gaming mouse
Keyboard Dell
Software Windows 10 Pro
From 25 mins in... that says it all. Storm-chaser, I implore you to have this on and listen to it.. if only from ~25 minutes in to the end.

"it's not that the website is now bad (because of its changes)... it was never good".

'should not be a resource for buying decisions...it's just not accurate'

"our audience isn't the audience that uses userbenchmarks.com" (inferring that enthusiasts are aware this isn't a good test but is for 'potatos')

I forgot GamersNexus also ripped into this site too.
https://www.reddit.com/r/buildapc/comments/chco8h
Knowing is half the battle, ehh? Otherwise it's just red and blue lasers (looks at my avatar)....

They tainted it. I told you this benchmark was shyte! I didn't know about the changes they made last year that made it worse. Respectfully, I agree with HW Unboxed, GN, and many others in the know. I'm out. GL to you, SC. :)

Okay, thanks EarthDog, wonderful contributions to the thread, I might add. But Seriously. If I cared what Gamers'Nexus and you thought of userbenchmark.com I wouldn't have started this thread in the first place... okay?
 
Joined
Jul 14, 2006
Messages
2,405 (0.37/day)
Location
People's Republic of America
System Name It's just a computer
Processor i9-9900K Direct Die
Motherboard eVGA Z390 Dark
Cooling Dual D5T Vario, XSPC BayRes, Nemesis GTR560, NF-A14-iPPC3000PWM, NF-A14-iPPC2000, HK IV Pro Nickel
Memory G.Skill F4-4500C19D-16GTZKKE or G.Skill F4-3600C16D-16GTZ or G.Skill F4-4000C19D-32GTZSW
Video Card(s) eVGA RTX2080 FTW3 Ultra
Storage Samsung 960 EVO M.2
Display(s) LG 32GK650F
Case Thermaltake Xaser VI
Audio Device(s) Auzentech X-Meridian 7.1 2G/Z-5500
Power Supply Seasonic Prime PX-1300
Mouse Logitech
Keyboard Logitech
Software Win7 Ultimate x64 SP1
Joined
May 6, 2018
Messages
1,150 (0.53/day)
Location
Upstate NY
System Name Dual Socket HP z820 Workstation
Processor Twin Intel Xeon E5 2673 v2 OEM processors (thats a total of 16C/32T)
Motherboard HP Dual Socket Motherboard
Cooling Stock HP liquid cooling
Memory 64GB Registered ECC memory kit (octal channel memory on this rig)
Video Card(s) MSI RX 5700 XT Gaming X 8GB
Storage 2 x 512GB SSD in raid 0
Display(s) Acer 23" 75Hz Gaming monitors 1080P x2
Case Brushed Aluminium
Audio Device(s) Integrated (5.1)
Power Supply HP 1125W Stock PSU
Mouse gaming mouse
Keyboard Dell
Software Windows 10 Pro
Top