Overclock.net banner
1 - 15 of 15 Posts

·
Twin Turbski
Joined
·
522 Posts
Discussion Starter · #1 ·
Forgive the 500 page book I've written below.

I know enough to know that the majority of the computing enthusiasts out there detest this benchmark and claim massive bias and rigged results designed to get you to buy certain hardware based on one or two YouTube videos that are always re-posted whenever someone brings up the subject of UBM as evidence of bias... But please, I just said it, so it doesn't need to be re-hashed again and again. We know that many think it's a controversial benchmark. Lets move on from there... the real question is do you think we can still find a way to make this benchmark viable for competition?

Reason being, is it's a very quick benchmark but also tests a much broader range of hardware than most benchmark engines. The results pool from userbenchmark.com is very extensive. In other words, we are dealing with a very large sample size, regardless of whatever hardware you have. Meaning, we will see a fairly accurate depiction of how your hardware performs against the UBM database. Essentially, you will be competing against all other submissions in the UBM database. Meaning you will not specifically be competing against other members here at overclock.net, more, you will be competing against similar hardware benchmarks submitted to UBM. That being said, there will ALSO be room to compare member to member results as well, if we get enough interest. Or grudge racing, if you want.

Since we all play by the same rules and all run the same UBM bench engine, your result can confidently be measured against the competition with NO doubts that the benchmark itself is skewing results. An accurate and legitimate comparison can then be drawn against your competition. And you can be confident in your result!

This means your results against like hardware will be legitimate.

I want to make one thing crystal clear. I am not in love with userbenchmark, nor do they sponsor me. I am in no way connected to them. I am not "raving" about userbenchmark.com nor am I being a "sold out fanboy" who can do no wrong -- I'm merely indifferent to the rumors/delusions leveled against userbenchmark.com because I see it much differently than you do. I see intrinsic value in it. I see the benchmark engine as a tool for measuring or estimating system performance under specific circumstances and If you are not ambitious enough to see a potential usage case here then I suppose there is no way to out reason your conclusions. Bottom line? Until you see the value in this benchmark, you will be forever stuck on that perpetual hamster wheel, repeating the same negative lines over and over and over again, sounding like a broken record while at the same time, desperately fighting to change someone else's benchmark competition because it doesnt meet your standards of "purity". Hint: Read that last sentence again.


Once all the naysayers stopped posting, and after I lost all credibility for hosting it (lol) we actually had a very successful UBM competition at techpowerup last year. I am going to show you the leaderboard so you can see the details on how I am assessing your submissions. By no means is this set in stone. And please, feel free to assess my leaderboard and tell us if you have an idea or a better way to assess your results. So we can do it like this or switch things up a bit, it's all about what the members here want to do.

All I am saying is we don't have to throw the baby out with the bathwater.... and FYI we had an excellent turnout for this comp once the haters stopped whining.

All ideas welcome, don't be afraid to be critical or offer your own opinion about how we can do this better...

Font Parallel Screenshot Circle Slope

Font Material property Pattern Parallel Rectangle
 

·
Registered
Joined
·
49 Posts
okay fanboy...... j/k that being said I honestly have no idea, I have used userbenchmark for years the problem is the bias ofc, but on top of that they dont use real world work loads, If we were going to do any sorta benchamrking why dont we use real world applications like games for gpus, we can pick free ones such as COD WZ or whatever and can use cpu workloads like video encoding sofrware or whatever the current one requires. I think that would be a better way to make it as un bias as well as even as possible. For gpu workloads we can measure in fps along with in ms for cpu we can calculate via time taken to render or compress or compile whatever
 

·
Twin Turbski
Joined
·
522 Posts
Discussion Starter · #4 ·
I somehow have the 3rd highest score ever for a 5960X and that was at only 4.5GHz. I should really run it at 4.7 just so I can grab the top score and be done with it.

Asus X99-A/USB 3.1 Performance Results - UserBenchmark
I hope you don't mind me posting a screenshot of your results. I am using this because it's a perfect example of how we could possibly make this viable. In overall percentage score. And the person with the highest score overall goes to the top of the leaderboard. That does not mean we have to disregard other aspects of the UBM, for example, you have an exceptional CPU score, and if you look at the sample leaderboard above you can see how we can integrate all the other hardware into your submission.

Font Terrestrial plant Screenshot Number Document
 

·
Twin Turbski
Joined
·
522 Posts
Discussion Starter · #5 ·

Attachments

·
Registered
Joined
·
49 Posts

·
Premium Member
Joined
·
517 Posts
I dont have any, my mb is a asrock z490 Extreme 4, and I only use the drivers I dont even use asrocks rgb software so im stuck with rainbow puke
I also have an MSI MB and it is installed by default/Windows.

Look at your MB ASRock Z490 Extreme4 support page and Nahimic is listed. My Asus MB also install it by default.

But maybe you system didn't install it if so, nevermind :)
 

Attachments

1 - 15 of 15 Posts
Top