Discussion Should we feel guilty that websites such as "Userbenchmark" are seen by millions of uninformed users?

I am not sure if most of you are familiar with user benchmark. It's one of the most popular sites for CPU comparisons and will come first in Google search with keywords such as "Ryzen 3600 vs Intel 9600KF". However, this is one of the most misleading sites I've ever came across to. If you go on their about page, they say:

"Marketers make it tough to choose hardware. A corporate army of fake forum and reddit accounts spread hype and disinformation to drive sales. Incompetent (moar core) smearers would sell ice to Elsa."

:SMH: They're somehow against the idea of CPU makers selling more cores to the average consumers, with its obvious benefits being more and more proven by sites such as this site, AnandTech, and many more. They even consider the TR 3970X to have the same "effective speed" as the 9600KF! Oh dear.

Whatever the reason of this madness is, I won't judge. But, we as enthusiasts must not let this happen, right? Should we start a campaign against this misinformation the way we campaign to never use cheapo PSUs? What are your thoughts?
 
It's useful for comparing your system to itself when you're tweaking settings.

It's also useful for comparing systems to others of identical architecture (same CPU, same or same type GPU, memory, etc.) but comparing across platforms is meaningless.

But isn't that true of a lot of benchmarks? There are reams of different benchmarks used to evaluate servers, for instance. You can't compare one to another, or one to another server when they're optimized to do a different job.

EDIT...and actually, thinking about it, it's probably useless comparing to even identical systems. Just like 3DMark is for comparing GPU performance. The avereages are so heavily skewed by the extreme benchmark competitors who have dozens upon dozens of runs with identical hardware as they tweak their overclocks. You're never comparing to the real, stable, 24/7 capable systems like average users want.
 
Last edited:

USAFRet

Titan
Moderator
Indeed. We see far too many people here trying to "fix" something that is not broken. Simply from "Way Below Expectations"
When in reality, their thing is performing near perfect.

Like this one:
 
This is why I take all of these sites with a grain of salt. The good thing is that there are plenty to use in tandem to see if a true variance exists.

Like how the i5-680 actually is faster across the board compare to the rest of the i5 in that same generation. No matter what site you check out, they all show the same relative result.
 
Last edited:
Might as well shut down YouTube as well. The Web is Vast! The Web is Mighty! And only about 10% truth.[/sarcasm]

Too much ado about nothing. I wouldn't worry because you know what the universe does when you try to make something idiot-proof.

-Wolf sends
We're talking about something that comes consistently on top of Google search results and is trying to spread a completely wrong message on buying advice. Not just wrong, but absolutely wrong.


Interesting. Fun fact, this is the guy said to be one of the "Incompetent (moar core) smearers" by Userbenchmark.
 

EndEffeKt_24

Commendable
Mar 27, 2019
659
157
1,340
I saw that clip from hardwareunboxed too. Damn it who the hell at userbenchmark is driving a campain versus AMD?
In the "conclusion" on the 3600 and 3700x they advise to buy a 9400k and 9600k respectively because it would be faster in gaming and streaming.
Doesnt matter what you look up on that site it always tells you to buy intel.
 
there ARE some very rational and reliable YouTube reviewers. Be careful about throwing the baby out with the bath.

I know of about two. Steve from GamersNexus and Aris from misc. sites doing PSU reviews. Beyond that, it starts getting rather thin, rather quickly. The majority of them give you quick tests, quicker screens, and assume you should believe what they say because "charts". Pffft. Right. There is practically nothing on Youtube that can't be faked. In fact, there is practically nothing that HASN'T been faked on there. LOL.

Sure, there are a few that are solid but most of it is highly skeptical, at best.
 
I know of about two. Steve from GamersNexus and Aris from misc. sites doing PSU reviews.

There's a lot more than that. Buildzoid, PC Games Hardware DE, Hardware Unboxed, Digital Foundry, mentioning a few here. These channels at least provide as useful information as written sites such as this one, if not more. Certainly not only "give you quick tests, quicker screens, and assume you should believe what they say because "charts"", let alone fake (seriously?). I'm getting too far from the discussion, though.

While the * vs * from Userbenchmark can be somewhat misleading I still think it's good for diagnosing issues such as memory not running at the correct speeds with Ryzen causing poor performance/dips or high CPU usage, etc.

Sure, but that wasn't my concern. My concern was about the buying advice, and it's not just "somewhat misleading", it's VERY misleading. 9600KF and 3970X has equal "effective performance"? That's equivalent of a very prominent site saying "CX550 performs the same as an AX1000 effectively".

Doesnt matter what you look up on that site it always tells you to buy intel.

I don't endorse conspiracy theories like that either. Regardless, the site's buying advice is major misinformation.
 
Last edited:

Karadjgne

Titan
Ambassador
Everything must be seen in perspective. And that's the biggest issue I see, the total lack thereof. It's quite easy to say that for instance the i5 6400 is better than the i3 6100. Realistically the i3 destroys the i5 6400 in everything except heavily loaded 4 core apps, even ties with the i5 6500 in some things. Perspective. Without it, you don't get the big picture and an overall sense of what's what.

I'm sure the 9600kf is comparable to the 3970X. In one bench run, of 1 specified test, with one certain pc, running one certain OS setup. And that's the part userbenchmark fails at, just what's compared, at what settings, with what retarded software running etc etc.
 
There's a lot more than that. Buildzoid, PC Games Hardware DE, Hardware Unboxed, Digital Foundry, mentioning a few here. These channels at least provide as useful information as written sites such as this one, if not more. Certainly not only "give you quick tests, quicker screens, and assume you should believe what they say because "charts"", let alone fake (seriously?). I'm getting too far from the discussion, though.
This doesn't even make sense. At all. In any way.
 
This doesn't even make sense. At all. In any way.
It's not complicated. You seem to be not familiar with YouTube. I don't want to talk about this topic any further though, it's not a relevant discussion to the thread.

I'm sure the 9600kf is comparable to the 3970X. In one bench run, of 1 specified test, with one certain pc, running one certain OS setup. And that's the part userbenchmark fails at, just what's compared, at what settings, with what retarded software running etc etc.

For buying advice, this is a huge problem. I think we can all agree on that. The question is should we care enough to tackle this?
 
Last edited:
That's truth. I don't care who the evidence is from, it should be supported by more than one source or it's unsubstantiated, even when it's somebody you trust. Every reviewer out there, and every enthusiast who thinks they are a reviewer, is capable of making mistakes with testing or with the test data. And, regardless that there are so many sheep who apparently believe everything that their favorite personalities suggest on Youtube or places like it, there are a lot of them that are simply unfit for what they are doing or are outright feeding people fake data. It's been proven time and time again. So whether you believe it or not is irrelevant. It IS what it IS.