[SOLVED] My CPU has degraded quite a lot in the past 5 weeks

Jul 19, 2019
13
1
15
Solution
The immediate thought is Userbenchmark changed its weighting system to assess the CPU's score. So in some cases what was a good score might be lowered because they've changed certain aspects. Your SSD and RAM scores are supposedly performing below spec.

And don't forget, Userbenchmark is a user aggregate site comparing scores with other people's system. It isn't a real benchmark where all factors are under controlled conditions for benchmark comparison between two specific components.
The immediate thought is Userbenchmark changed its weighting system to assess the CPU's score. So in some cases what was a good score might be lowered because they've changed certain aspects. Your SSD and RAM scores are supposedly performing below spec.

And don't forget, Userbenchmark is a user aggregate site comparing scores with other people's system. It isn't a real benchmark where all factors are under controlled conditions for benchmark comparison between two specific components.
 
Solution
Except those weighting changes were aimed at lowering AMD's Ryzen 3000 scores and making Intel CPU's appear more competitive. If OP has an Intel CPU... that makes no sense.

So the conspiracy interpretation goes. Nevertheless, for any CPU if the weighting is changed then the score would change too (all else being the same); whether it changes for the better or worse is dependent upon the CPU. What is true, what you allude to, is one would expect quad cores to become better CPUs according to Userbenchmark.
 
So the conspiracy interpretation goes. Nevertheless, for any CPU if the weighting is changed then the score would change too (all else being the same); whether it changes for the better or worse is dependent upon the CPU. What is true, what you allude to, is one would expect quad cores to become better CPUs according to Userbenchmark.

I don't really think it's a conspiracy theory since the folks at Userbench even said as such, and that they said it was 'needed'.
https://www.techspot.com/news/81176-userbenchmark-offers-explanation-changes-cpu-score-weights.html

And from what I've read, the changes were only in how they weighted scores of Ry3k CPU's and no other CPU's. SO...comparing a raw score from 'then' to a raw score from 'now', for a CPU that hasn't had it's weighting altered, might be valid. If it went down, maybe there is something to it.

And maybe not. Userbenchmark is a pretty bad way to measure system performance, that's for certain, since by their own admission they apply weighting factors to force various CPU's to perform according to how they feel they should perform.
 
Don't disagree, and aware of the situation.

Though from what I've read and heard, the weighting was a general change rather than specific to a specific line of CPUs. A weighting change which would favour a different set of CPUs. If they only changed the weighting for a specific set of CPUs, then it would just be plainly discrimatory. I wouldn't rely on Userbenchmark for any serious bench testing because it is user to user with no controlled conditions so offers very little in the way of 'real comparison'. Gamers Nexus slams it as effectively useless, which I don't disagree with either.

However, none of this helps OP.
 
Well, no.
But in any appreciable user timeline...they don't.
Ahh...so we've established that much at least :)

But I think overclocking the snot out of a processor can have the effect of shortening that timeline WAYYYyyyy down for it as I've actually had it happen to me. I'd overclocked a 386sx-25 to something like 50Mhz (don't take it for gospel...i really can't remember THAT detail). It ran my d-base indexing tasks for several months but then it started throwing errors in the indexes and finally crashing. Processors ran wide-open all the time in those days; it would crash just sitting at the DOS prompt (I'd go to type a command and nothing would happen, not even an echo-back of keystrokes). I opened it up and moved some jumpers around to get clock back down and it booted and ran fine then. I may be naive, but I'd say I had 'degraded' my processor.

I managed to swing budget for an upgrade soon after but who's to say, after my abuse, it just might have continued right on...ummm, degrading let's say...until the poor sot who got it gave up and tossed it into the trash.

Whether that is happening to OP or not I can't say...it might also depend on how Intel processor's work internally. If 'degrading' should take the form of not boosting as much as it did before then performance could suffer without the crashes and erroring I had. I don't know that's his case, I'm just saying I see a way for processors to 'degrade' without waiting for the universe to cool down to absolute zero.
 
Last edited: