I9 9900k

[moorhen2]

That's at 1.25v?

Yes that is correct, under load it was 1.219v.

 

[moorhen2]

What LLC?

LLC of 7

 

[Nickolp1974]

LLC of 7

Thanks, actual bit more than 1.25v then

 

[Distracted]

Thanks, actual bit more than 1.25v then

Showing as less for me with ll7.

Setting 1.3 in bios with ll7 gives 1.288 max in hwinfo.

I think gamers nexus saw similar droop.

 

[moorhen2]

Thanks, actual bit more than 1.25v then

1.260v in bios, but all monitoring software state 1.250v, ie HWinfo, CPU-Z etc.

But states 1.219v under full load in Realbench and Cinebench.

 

[Sc00p007]

Yes that is correct, under load it was 1.219v.

Wow, nice. 1.219v at full load for 5.1ghz? Fully stable with no throttling?

 

[Distracted]

1.260v in bios, but all monitoring software state 1.250v, ie HWinfo, CPU-Z etc.

But states 1.219v under full load in Realbench and Cinebench.

Tyvm for sharing.

I think there is a large variance in the quality of the 9900k cpus that are out there.

 

[moorhen2]

Wow, nice. 1.219v at full load for 5.1ghz? Fully stable with no throttling?

Not saying its fully stable, 30 minute run of Realbench Stress test doesn't mean much, but does give me a little pointer as to what this chip might be capable of.

 

[moorhen2]

Tyvm for sharing.

I think there is a large variance in the quality of the 9900k cpus that are out there.

Yes its the old Silicon Lottery, I purchased this 9900k from OC's on Monday, its a tray or OEM whatever you want to call it.

 

[Distracted]

Is realbench considered a realistic test of the system for determining day to day stability?

I can see from using it that it has some AVX stuff in there but also stresses components other than the CPU.

Note to self. Stop touching the mouse.

 

[moorhen2]

Is realbench considered a realistic test of the system for determining day to day stability?

I can see from using it that it has some AVX stuff in there but also stresses components other than the CPU.

Note to self. Stop touching the mouse.

If it stops when you move the mouse your only running the Benchmark, try the stress test.

 

[Distracted]

If it stops when you move the mouse your only running the Benchmark, try the stress test.

Yeah, I was just running the benchmark because I have a few minutes. I won't stress it until later today/tomorrow.

I'm just wondering when I can consider the system stable and what programs to use for stressing it. I'll just add realbench to the steadily growing list.

 

[moorhen2]

Yeah, I was just running the benchmark because I have a few minutes. I won't stress it until later today/tomorrow.

I'm just wondering when I can consider the system stable and what programs to use for stressing it. I'll just add realbench to the steadily growing list.

Well I don't and never have used P95, some still bang on about it, but its personal choice, on another forum I belong to there was a stability thread and they advocated 8 hour runs of the Realbench Stress test, which I did do with my other chips, but I don't think is really needed. I am an enthusiast, I love pushing hardware, but I am also a great believer in that if your PC runs and does what you need it to do day in and day out without crashing that's good enough, but that is just me.

 

[0121danwilliams84]

Tyvm for sharing.

I think there is a large variance in the quality of the 9900k cpus that are out there.

Tell me about it. 1.320v for 4.9Ghz here!

 

[Distracted]

Tell me about it. 1.320v for 4.9Ghz here!

I don't know what is the norm right now. Hopefully the binning stats silicon lottery release on sat will reveal a bit more info.

 

[0121danwilliams84]

I don't know what is the norm right now. Hopefully the binning stats silicon lottery release on sat will reveal a bit more info.

Oooh shall definitely be looking into that TY

 

[moorhen2]

I don't know what is the norm right now. Hopefully the binning stats silicon lottery release on sat will reveal a bit more info.

That's ok, but it will only give a tiny percentage of silicon expectancy, so probably wont be a real indicator, don't forget that the batches of silicon produced will mature over time, but as always it will be a lottery. Some say batch numbers can be an indicator of possible performance, ie where on the wafer the chip came from etc etc. I know my chip was produced in week 35 of this year and was number 80 on the wafer, does this indicate that it will perform better than another, who knows.

 

[0121danwilliams84]

That's ok, but it will only give a tiny percentage of silicon expectancy, so probably wont be a real indicator, don't forget that the batches of silicon produced will mature over time, but as always it will be a lottery. Some say batch numbers can be an indicator of possible performance, ie where on the wafer the chip came from etc etc. I know my chip was produced in week 35 of this year and was number 80 on the wafer, does this indicate that it will perform better than another, who knows.

That was the reported golden sample!

 

[moorhen2]

Up to 5.2gh now, just testing core multiplier for now, booted into the OS without touching vcore, tried a quick run of Cinebench and only got half way through before it stopped, not a hard lock, which would indicate cache, but stopped running. So 0.5v added to vcore and run again and passed. Not an indicator in any way, but a step in the right direction. 1.225v under load.

 

[0121danwilliams84]

Up to 5.2gh now, just testing core multiplier for now, booted into the OS without touching vcore, tried a quick run of Cinebench and only got half way through before it stopped, not a hard lock, which would indicate cache, but stopped running. So 0.5v added to vcore and run again and passed. Not an indicator in any way, but a step in the right direction. 1.225v under load.

Looking like the makings of a good chip there buddy