R6E: Any clue on what is the expected i9-7940X overclock?

[Inkar]

Hi,

Does anyone have a clue about what is the all core expected i9-7940X (non-avx) overclock range with the Rampage VI Extreme and an AIO?.
I know nothing is guaranteed other than the factory specs. I just can't find any reviews with this kind of data for the 7940X.


Thanks.

[Silent_Scone]

Protocol wrote:
No "forgiving" testing for me. If it won't run any instruction set in any stress program/benchmark/non-synthetic workload at absolute 100 percent for at least 8 hours then I do not consider my system stable. (I only perform 8-hour testing if a smaller 2-hour test passes. Usually minor tweaks over a day or so gets everything dialed in tightly). Being lidded, I am hitting a thermal wall myself. Just can't get enough heat out and into the waterblock/coolant.

Skylake-X in general suffers from heat-transference issues due to the TIM compound that is used. I'm sure a hefty number of people are able to hit clocks with decent voltage overhead remaining but are reaching the TMAX instead (non-delidded of course).

This methodology doesn't work well on these platforms, unless you're ok with sacrificing performance. Nobody really wants to do that, do they? There is little to no point at all subjecting the CPU to that much torture if you don't plan on using AVX workloads.

[Protocol]

Yes, very aware of this. Works fine on mine. Highly subjective really because of course it’s going to depend entirely on what "the goal" is! Having an 8-hour render fail due to that extra 100 MHz that tips the scales doesn’t make sense for some. I do a lot of HEVC encoding and other parallel workloads. AVX2 primarily as I tend to soft-disable AVX-512, for now.

If anything one would want to test more throughly with AVX workloads given the characteristic power/heat tendencies.

Sacraficing performance is subjective because performance itself is subjective. To me performance is stability at decent clocks and temperatures for an extended amount of time. To others they might be content with the opposite if their workload supports it.

[Silent_Scone]

Protocol wrote:
Yes, very aware of this. Works fine on mine. Highly subjective really because of course it’s going to depend entirely on what "the goal" is! Having an 8-hour render fail due to that extra 100 MHz that tips the scales doesn’t make sense for some. I do a lot of HEVC encoding and other parallel workloads. AVX2 primarily as I tend to soft-disable AVX-512, for now.

If anything one would want to test more throughly with AVX workloads given the characteristic power/heat tendencies.

Sacraficing performance is subjective because performance itself is subjective. To me performance is stability at decent clocks and temperatures for an extended amount of time. To others they might be content with the opposite if their workload supports it.

When talking about overclocking, the metric is frequency for a given workload. As per my previous post:

Silent Scone wrote:
There is little to no point at all subjecting the CPU to that much torture if you don't plan on using AVX workloads.

If the stress tests are relevant to a given workload, then that's fine. That's a little different from saying the below... 😉

Protocol wrote:
No "forgiving" testing for me. If it won't run any instruction set in any stress program/benchmark/non-synthetic workload at absolute 100 percent for at least 8 hours then I do not consider my system stable.

All the more reason to make good use of AVX and AVX512 offsets.

[Hot_Trout]

Well I have been for the past few weeks running 24/7 a mixture of large scale zip work consisting of over 42,000 files being opened, changed and repackaged from one SSD to another using all cores. This produces heavy loads on all cores, heavy memory use and constant I/O. This has been stable. The PC also runs blender for 6+ hours with 100% CPU without fail. I have not de-lidded but I did use liquid metal paste between the CPU and the HUGE dedicated EK CPU & VRM waterblock.

Can I just ask again what people with the 7940x are setting their VCCIO and CPU System Agent to?

[Marko911]

I Oced mine i9 7900x stock TIM,I couldn't pass PRime95 26,yesterday I did delidding with rockit99 and my temp in idle are 26,full load 75C...So I don't think I could expect something more from OC beyond 4,0 all cores expect little better temps at full load.Will give full report in a few days.Hope it helps.

[Silent_Scone]

Marko911 wrote:
I Oced mine i9 7900x stock TIM,I couldn't pass PRime95 26,yesterday I did delidding with rockit99 and my temp in idle are 26,full load 75C...So I don't think I could expect something more from OC beyond 4,0 all cores expect little better temps at full load.Will give full report in a few days.Hope it helps.

Did you buy the CPU to run Prime all day?

Hot Trout wrote:
Well I have been for the past few weeks running 24/7 a mixture of large scale zip work consisting of over 42,000 files being opened, changed and repackaged from one SSD to another using all cores. This produces heavy loads on all cores, heavy memory use and constant I/O. This has been stable. The PC also runs blender for 6+ hours with 100% CPU without fail. I have not de-lidded but I did use liquid metal paste between the CPU and the HUGE dedicated EK CPU & VRM waterblock.

Can I just ask again what people with the 7940x are setting their VCCIO and CPU System Agent to?

CPU and memory kit dependent, but for 4000Mhz 4x8GB I'm using VCCSA 0.880v and VCCIO 1V

[Marko911]

Silent Scone wrote:
Did you buy the CPU to run Prime all day?



CPU and memory kit dependent, but for 4000Mhz 4x8GB I'm using VCCSA 0.880v and VCCIO 1V

Point taken,I'm new to HEDT platform so I'm used to run prime on i7 7700k...Can you suggest me something to test stability on i9?I didn't have any issue with TPU II OC profile but,but temps on those settings are pretty much high for delid CPU...

[Silent_Scone]

Marko911 wrote:
Point taken,I'm new to HEDT platform so I'm used to run prime on i7 7700k...Can you suggest me something to test stability on i9?I didn't have any issue with TPU II OC profile but,but temps on those settings are pretty much high for delid CPU...

The point was really just that, it depends on the workload. So what is your workload? 😉

[Marko911]

Silent Scone wrote:
The point was really just that, it depends on the workload. So what is your workload? 😉

I bought it so I could do a faster job on my current project that involves 3D rendering (and yes I suck at it 😄 )
and I'm in process of expanding m.2 storage to at least 2TB more.-there is also a high daily data transfers on my PC (call me crazy but SATA SSD and HDD are for me way,way slow)
Gaming wise I didn't noticed any performance impact ,CPU barely goes above 45C .

[Hot_Trout]

Marko911 wrote:
I bought it so I could do a faster job on my current project that involves 3D rendering (and yes I suck at it 😄 )
and I'm in process of expanding m.2 storage to at least 2TB more.-there is also a high daily data transfers on my PC (call me crazy but SATA SSD and HDD are for me way,way slow)
Gaming wise I didn't noticed any performance impact ,CPU barely goes above 45C .

Similar requirements here. For my motherboard I choose to put the OS and Apps onto an Optane 900p riser card (fastest storage around). That left me both the M.2 slots for dedicated Samsung 960 Pro drives. I can average somewhere around 2500MB/s between disks during repacking. I noticed that the i9 released some lost performance from within my 980ti but the real benefit was being able to multitask without any slow down.