*NEW UPDATED 2* Post Your Cinebench R11.5 Score

[ericlp]

small bump OC for my TR running on Air...

 

[crashtech]

This chip doesn't clock high, but it stays amazingly cool for some reason, even with just a Hyper 212, 57° at full load. The 1700X I picked up early on runs a LOT hotter than this chip. Voltage was 1.27, btw, the screencap didn't catch it.

 

[Bushwicktrini]

Here's Mine didn't take a screenshot of 14.82

https://imgur.com/84BV4gU

 

[crashtech]

@Bushwicktrini , I can't include your entry as yet, please review the OP for details.

 

[Totalnet]

My submission for the i5 (4t) class
I was testing with LN2..

 

[Totalnet]

Another LN2 test, 7820X performing very well..

 

[Totalnet]

He guys, finaly tried my 7980XE on LN2
Temp about -100 degrees (celsius) to prevent cold-bug
This chip needs a lot of LN2 and power..

 

[crashtech]

@Totalnet , I'm going to leave both of your 7980XE scores in place, since it's the only sample of its type on the list. I'm a bit curious to know what sort of speeds one could expect from your CPU with something like a custom loop with a large radiator instead of the chiller.

 

[Totalnet]

@Totalnet , I'm going to leave both of your 7980XE scores in place, since it's the only sample of its type on the list. I'm a bit curious to know what sort of speeds one could expect from your CPU with something like a custom loop with a large radiator instead of the chiller.

Result today on a hot day (28 degrees)
system :
1x rad 560 (4x 140mm fan) + 1x rad 320 (3x 120mm fan ) both push configuration (see pic)
The rest is the same config
CPU 7980XE (not delidded)

 

[Markfw]

@Totalnet why does it say 16 core, not 18 core ? on the cinebench screen

 

[crashtech]

@Totalnet , thanks so much for doing that! The results show a fairly linear progression, and illustrate the extra performance users can get for the extra effort and money. While the results with more esoteric cooling are astounding, the regular* watercooled result is not bad at all! I would guess that many users with the money to spend would still opt for that somewhat simpler solution.

*Well, not really "regular." That's a whole lot of cooling area!!

 

[Totalnet]

@Totalnet why does it say 16 core, not 18 core ? on the cinebench screen

@Markfw Cinebench 11.5 is a very old benchmark and it can only report 16 cores per CPU

@crashtech
This amount of cooling surface can be found in a big case, i have seen a lot of builds with 2x 560 rads (long version) in the back of the case with a push/pull config, so it's not realy regular but possible..lol
This CPU does not need a lot of volts for this performance, that helps a lot

 

[crashtech]

Yes, R11.5 is very old! But I can't just abandon all the data. It may be that at some point this thread will die a quiet death. Perhaps I should start another Cinebench R15 thread, because the other one is not maintained, at least not how I would want it to be.

 

[Totalnet]

@crashtech i think you are a good and responsive mod and there are a lot of fun benchmarks
So if you want to start something i'm in ..

 

[crashtech]

I'm flattered, let's see if we get a positive response from some other members. I'm not a mod, by the way, but thanks for the recommendation!

 

[mjdupuis]

Scores for both 4.8 (my daily driver) and 5GHz. I'm not delidded so I don't really use my 5.0 profile as the temps get a little too high, Cinebench got the temp on one core to 83C. It's stable but I'll stick with 4.8 until I decide to delid.

 

[crashtech]

@mjdupuis , your setup seems a bit ahead of the pack per clock, once delidded, it would not be surprising if you took top hexacore honors.

 

[mjdupuis]

@mjdupuis , your setup seems a bit ahead of the pack per clock, once delidded, it would not be surprising if you took top hexacore honors.

I think it may have to do with memory speed, 3400MHz C14 with tight timings.

 

[Justinus]

I think it may have to do with memory speed, 3400MHz C14 with tight timings.

You make me want to do something better with my 3600CL15 memory than running it at 3200 on a 7600k

 

[crashtech]

Another cheap DC rig:

 

[JoeRambo]

I think it may have to do with memory speed, 3400MHz C14 with tight timings.

Yeah, 4.8/4.4 uncore 8700K with 3200CL13 seems to have even better score:

 

[Anubis]

delidded and pushed it to 5GHz - back in the top 5 woo woo

no temps is that but this was right before 2 hours into prime

 

[JoeRambo]

Decided to do a 5Ghz run with memory as tight as in 4.8Ghz run. Not a daily driver, cause voltage required is ~1.385V and temps rise too high during OCCT runs. Still cool to match 5.1Ghz system with 100mhz deficit.