Edrick, do you have Geekbench 4? If you have it, I would be very interested in seeing a run at 2400MHz L3 and then one at 3200MHz L3, then we can get some insight into how the faster L3 impacts CPU-focused benchmarks.
All the reviews of skylake-x I have seen have been tested with stock 2400mhz northbridge. I have shown that just upping that to 3200mhz shows a measurable gain. I wonder when we will see reviews with this value overclocked.
All you need is active cooling on the VRMs.
My Z170 Asrock Pro4s throttled my OC'd 7700k - a fan over the VRM heat sink sorted it.
Airflow solves the VRM heat issue, AVX offset keeps power under 270W to solve the connector/wiring heat issue.While this will solve the VRM heat issue, your 7700k draws a lot less power from the 8 pin power connector.
Edrick, do you have Geekbench 4? If you have it, I would be very interested in seeing a run at 2400MHz L3 and then one at 3200MHz L3, then we can get some insight into how the faster L3 impacts CPU-focused benchmarks.
Yeah, that was what i was looking for, old stuff with inclusive L3 had 80 and up depending on ring size in that test. Intel now has some real nice latency for some subtle thread synch/communication cases.
So gaming problems are most likely to be memory latency and uncore speed/latency related.
I'd love to see 3200 uncore and CL14 3200 mem run compared to default settings in gaming tests that "suffer".
I'm now wondering what exactly PCPer was measuring when they said core to core was 100ns.
Yeah... But it could be that mlc does not support Skylake-X yet and is not measuring correctly. I suspect that to be the case, cause L2 HITM should be slower, after all it is hitting L2 line that is owned and modified by other core. Makes little sense when it is faster.
But it could also be that pcper is eating memory write + read from say eviction L3 or sth, without seeing code and CPU counters it is hard to say exactly how they are testing.
I still have no idea what is going on with the voltage issue I am having. For the first few hours everything was working fine. I was able to set the voltage and clock speed. HWMonitor and the bios were both showing the correct voltage as it had been set. At some point as I had been adjusting clocks and voltage I lost the ability to control voltage when demanding a set clock speed. For instance, when I set my CPU clock to 4.7GHz and the voltage to 1.225, it gives me a voltage of 1.273. If I change it to 1.3v, it gives me 1.3v. So, I can run a voltage higher, but not lower. I thought maybe it was as bios/motherboard issue, so I actually swapped out to another Aorus G7 I had...somehow I ended up with two...lol. Unfortunately the problem persists. Is it possible something happened to the CPU itself?
I never use fixed voltage except for when trying to find the stable OC. Once I find it, I go back in and get things running dynamically. I'll take a look in the bios again. The oddity is that it just started acting that way after behaving for a few hours.I seem to remember scratching my head over a similar issue with my rig. I do remember losing control over core voltage. It was dependent on how I was controlling the CPU. It was just BIOS settings relating to dynamic clocks, offset voltages etc. I doubt your CPU is broken. Try getting it to work with dynamic clocks and voltage where everything ramps up and down depending on load, assuming that's how you want your CPU to work. I can't stand it when people use fixed voltages personally. It bugs the absolute hell out of me. Why would anyone hold a torch to their CPU 24/7? Get those dynamic functions working for clocks and voltage using the offset voltage, at least that's what I would do.
I still have no idea what is going on with the voltage issue I am having. For the first few hours everything was working fine. I was able to set the voltage and clock speed. HWMonitor and the bios were both showing the correct voltage as it had been set. At some point as I had been adjusting clocks and voltage I lost the ability to control voltage when demanding a set clock speed. For instance, when I set my CPU clock to 4.7GHz and the voltage to 1.225, it gives me a voltage of 1.273. If I change it to 1.3v, it gives me 1.3v. So, I can run a voltage higher, but not lower. I thought maybe it was as bios/motherboard issue, so I actually swapped out to another Aorus G7 I had...somehow I ended up with two...lol. Unfortunately the problem persists. Is it possible something happened to the CPU itself?
I can not find SVID in the bios. This is crazy.Can you disable SVID in the BIOS?
With it being this way on two different boards and no one else having the issue, could i possibly be the chip? Can a chip dictate voltage in anyway? That is not possible right?
I don't think so, no. Honestly, these quirky BIOS is why I stay well away from Gigabyte boards - it's always something with them. I always put my money on Asus.
Did you try clearing the BIOS or maybe even re-flash it and start over?
If I was buying SKX, I'd go Asus ROG Rampage VI Apex. The VRM heatsinks on this board look better than the current crop.
https://www.asus.com/uk/Motherboards/ROG-RAMPAGE-VI-APEX/
Unfortunately yes. I re-flashed the bios and tried clearing the CMOS. I'll probably just rock it at 1.273v until the other ASUS offerings are available, then RMA this board and make the switch.
I was thinking about scooping up the Deluxe. I had a P8Z68 Deluxe that served me well. I may just wait and go for the Apex.Yeah do it. Get the Strix or a ROG board - even the prime deluxe.
The Apex is a pure extreme method OC board, lacking many features you'd welcome for normal usage, including quad channel RAM for example if I recall correctly. I'd pick a different board.
It supports quad channel (just not 8 DIMMs) and also includes all the typical features you'd expect onboard.
I wish I bought the Apex Z270 instead of the Code - i could have probably achieved better memory overclocks / stability while still retaining the necessary features.