Jump to content
HWBOT Community Forums

CPU benchmarks applicable for global points - 2019


Which CPU benchmarks should get global points?  

129 members have voted

  1. 1. Which CPU benchmarks should still get global points in 2019?


This poll is closed to new votes


Recommended Posts

5 hours ago, mickulty said:

It sounds counterintuitive that you can make your eyesight *better* by putting weird curvy bits of glass in front of your eyes as well, but you don't see me insisting to you that small books should just be got rid of and these "glasses" things are moronic.

Yes, there is an amount of stress that causes parts to fail.  A great way to stop them from failing is to limit the stress to below that amount.

Oh dear....

If you insist on a metaphor for some reason, this is akin to having a power-lifting competition and limiting the weight people can use because they might tear a muscle.

Thanks for trying to dumb it down to my level though. 

  • Like 2
  • Thanks 2
Link to comment
Share on other sites

10 hours ago, websmile said:

The by far best way to limit stress and stop hardware from failing is to stop overclocking and of course benching. Great proposal, but I am not sure this is the correct forum to post it..

troll_face.png

Well if you overclock you accept risk, that's what we tell people on day 1 isn't it?  But I don't want to tell people just straight up "wrong hobby" when they start complaining because there's a risk, better to work out a way to mitigate it if someone's going to be so concerned, people pick their own risk level.  Can't have it both ways, it doesn't make sense to go "my chip died bench bad" then "I don't wanna have limits, that's not the point".

Like... yeah, share your experiences.  Warn people.  Get upset, I would.  It sucks when hardware dies.  I just think it comes a long way below relevance, security and ability to run on different hardware when deciding if a benchmark is a good benchmark.

Off-topic but maybe relevant to some readers, apparently Luumi left his 9900K in a dewar for a couple of days and it's reanimated.

  • Like 3
Link to comment
Share on other sites

  • Administrators

You still don´t get the point - people take risks since 10 years like splave, but if there are no signs where the limits are at, you toss a coin. No one will be happy if cpus die randomly and limiting current or volts means you can skip the benchmark where you have to do it for obvious reasons, plus you have not even a guarantee this helps that the cpu survives. The discussion leads nowhere bc you practically say that you should drive a f1 at half speed bc engine can blow up if you go full out, nontheless some people will go full out because they want points in the ranking, especially people who dont care because they have es or sponsored cpus anyway

The most funny part of this discussion is that guys adress it which are in a priviledged position compared to normal user - on some of their cpus they do not even have to pay. About the relevance, how much more avx benchmarks that people at top flight do not bench if things go on the way they do you see as relevant? We had a major y cruncher desaster at wc finals, people nag about xtu all the time, now we see a very good benchmark like x265 under fire and even on some 3marks we now have avx at cpu tests according to benchers. We can include these all at hwbot, but we do not discuss about the benchmarks themselves here but about global points. As soon as you give global points you practically force people into doing a benchmark if they fight for ranking

  • Like 4
  • Confused 1
Link to comment
Share on other sites

I still think X265 is great benchmark for global, why AMD is fine when Intel is dying? we must blame AMD for making CPU too strong :p
BTW AVX is the future, if you want to compete on performance you must have strong AVX realibility too.

Ps : I like XTU, but the problem its Intel only and buggy score is annoying.

Edited by speed.fastest
  • Like 1
  • Thanks 1
  • Confused 1
Link to comment
Share on other sites

  • Administrators

AMD does not die because amd is obviously not capable of implementing avx correctly :D - if you look at x265 result, amd needs around 1ghz more for same score like Intel, while on cb15 for example these are on par per mhz. As soon as amd implements avx to full strength, you will see how strong their cpus really are^^

  • Like 1
  • Confused 1
Link to comment
Share on other sites

1 hour ago, websmile said:

AMD does not die because amd is obviously not capable of implementing avx correctly :D - if you look at x265 result, amd needs around 1ghz more for same score like Intel, while on cb15 for example these are on par per mhz. As soon as amd implements avx to full strength, you will see how strong their cpus really are^^

It's not just AVX, luumi had a 3770k die in the exact same way which does not have avx2. AVX units on ivy are very small compared to avx2 units on later arches.

  • Thanks 1
Link to comment
Share on other sites

  • Administrators

Yes, it is not only avx, if you delid the 3770k like Luumi did and put on insane volts, it can die as well - also on 32m or cpuz validation^^

P.S. Just asked Juhani - how do ypou come to the idea he ever ran x265 on 3770K? Fried at wprime at max volts^^

  • Like 2
Link to comment
Share on other sites

13 minutes ago, Splave said:

GPUPI with new Open CL is using AVX and not killing so its not just AVX to blame albeit x265 is still a greater load than gpupi so who knows...

Do you have GPUPI 4 for testing? Because I think current versions don't use it, at least not directly... and who knows what instructions is OCL driver using and how much. Also from my experience GPUPI (all public versions available at this time) is actually very light CPU benchmark, at lest on everything I ran it on.

  • Like 1
Link to comment
Share on other sites

1 minute ago, havli said:

Do you have GPUPI 4 for testing? Because I think current versions don't use it, at least not directly... and who knows what instructions is OCL driver using and how much. Also from my experience GPUPI (all public versions available at this time) is actually very light CPU benchmark, at lest on everything I ran it on.

Current version uses whatever opencl driver uses and the newest intel opencl driver uses avx.

  • Like 1
Link to comment
Share on other sites

22 minutes ago, speed.fastest said:

As earlier said, it was current kill your cpu, doesnt matter avx or not if it using so much current it will be killed. Solution is reducing your power consumption (can be vcore, pll termination, or something).

Or solution #2:  don't run this benchmark, which is the approach I'm going to take.

Link to comment
Share on other sites

16 minutes ago, yosarianilives said:

Current version uses whatever opencl driver uses and the newest intel opencl driver uses avx.

Ah, yes, that is possible. I almost forgot about Intel OCL driver as it was always slower for me. But is seems lot of people use it with modern CPUs. Anyway for example 9900k can run 100 MHz higher GPUPI 1B than R15, so even with Intel driver it is still rather light benchmark.

  • Like 1
Link to comment
Share on other sites

16 hours ago, speed.fastest said:

Im not forcing you to run, logic is if you run 2v/2.05v cinebench on 6700K/7700K its going to die soon.

That's not what we are doing. That's the problem. Reasonable voltage is being used so i dont see why everyone is offering foolish equivalency like this. 

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

On 12/10/2018 at 3:33 PM, havli said:

Ah, yes, that is possible. I almost forgot about Intel OCL driver as it was always slower for me. But is seems lot of people use it with modern CPUs. Anyway for example 9900k can run 100 MHz higher GPUPI 1B than R15, so even with Intel driver it is still rather light benchmark.

I think it's more relevant for Skylake-X as it also uses AVX-512

  • Like 1
Link to comment
Share on other sites

Btw since we're discussing benchmarks, what do you guys think about modern 3DMark CPU/Physics tests?

In my notes, Fire Strike physics doesn't show much scaling to DRAM Tweaks, but the Time Spy / Time Spy Extreme Physics scales a lot with memory performance(not just CPU clocks), these tests also showed up quite often of big International OC competitions, the scores also quite consistent and not much run variation like the old 3D11 Physics. IMHO these TS/TSE Physics would be a nice addition to the 2D benchmark we have now.

The downsides are Time Spy/Extreme test need an advanced key, some overclockers don't seem like to cash another couple bucks(even though they bought hardwares that costed many times the benchmark).

 



 

  • Like 7
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...