Just joined ... first post ever. Hope I put it in the right place.
Built my first gaming PC a month ago and have precisely 4 hrs experience OC'ing ... anything. I'm not including data so, yes, I'm hand waving.
I've got a 4.90GHz OC @ 1.300V (77C max) and it runs everything I normally do -- games being the most 'stressful' on the CPU. Can't get thru more than 45 sec's of OCCT AVX Capable Linpack without the processor shutting itself off.
After monitoring my 7700K's temp's, voltages and power dissipation during stress testing my gut instinct is telling me this: it's just physics.
I'm a EE with 40 years experience in board-level analog circuit design. I spent the last 5 years of my career working at one of the US National Laboratories run by MIT designing a first-of-its-kind free-space laser communication system. But I started my career as a power electronics engineer -- designed lots of DC power supplies and 1000V 100A motor drives for elevator systems, etc, etc. I used to be able to lick my thumb, put in on something hot and tell you the temp within 2-3 degrees. So I'll make the bold claim that my 'gut instinct' is somewhat educated re: power loads and things getting hot.
The volume and mass of the CPU silicon hasn't changed much from my old i5-4460 and the 7700K, but the max power under heavy load is easily 2.5x.
The i5-4460 can run the exact same OCCT and RealBench stress tests forever with NO cooling (comparatively) ... because the dissipation and current is so much less than the 7700K.
The 7700K has built-in over-temp, over-voltage and under-voltage safeguards. The most appealing notion is that it's just going over-temp and shutting down, but with 60-70A load current switching rapidly, it could be voltage transients tripping limits and causing shutdown.
I'm seeing a lot of enthusiasts making claims about faults in the 7700K re: OC'ing and temps, but I think it's just the physics of that much power load changing so rapidly (unnatural synthetic stress tests) in a tiny silicon mass.
Intel is advising customers to simply NOT overclock their CPU ...
If it IS the case that it's just physics, then Intel has hit a wall re: their enthusiast customer base -- at some power magnitude you just can't get the heat out well enough to maintain a live-able temp. I wouldn't expect an Intel engineer to state that; I'd expect to hear a marketing type say 'just don't overclock'.
Maybe I'm being tremendously obvious, but this is my opinion re: the tons of posts I'm seeing about the 7700K being somehow flawed and that's why the usual OC routines aren't successful. Maybe there IS a design flaw, but it could just be physics.
Built my first gaming PC a month ago and have precisely 4 hrs experience OC'ing ... anything. I'm not including data so, yes, I'm hand waving.
I've got a 4.90GHz OC @ 1.300V (77C max) and it runs everything I normally do -- games being the most 'stressful' on the CPU. Can't get thru more than 45 sec's of OCCT AVX Capable Linpack without the processor shutting itself off.
After monitoring my 7700K's temp's, voltages and power dissipation during stress testing my gut instinct is telling me this: it's just physics.
I'm a EE with 40 years experience in board-level analog circuit design. I spent the last 5 years of my career working at one of the US National Laboratories run by MIT designing a first-of-its-kind free-space laser communication system. But I started my career as a power electronics engineer -- designed lots of DC power supplies and 1000V 100A motor drives for elevator systems, etc, etc. I used to be able to lick my thumb, put in on something hot and tell you the temp within 2-3 degrees. So I'll make the bold claim that my 'gut instinct' is somewhat educated re: power loads and things getting hot.
The volume and mass of the CPU silicon hasn't changed much from my old i5-4460 and the 7700K, but the max power under heavy load is easily 2.5x.
The i5-4460 can run the exact same OCCT and RealBench stress tests forever with NO cooling (comparatively) ... because the dissipation and current is so much less than the 7700K.
The 7700K has built-in over-temp, over-voltage and under-voltage safeguards. The most appealing notion is that it's just going over-temp and shutting down, but with 60-70A load current switching rapidly, it could be voltage transients tripping limits and causing shutdown.
I'm seeing a lot of enthusiasts making claims about faults in the 7700K re: OC'ing and temps, but I think it's just the physics of that much power load changing so rapidly (unnatural synthetic stress tests) in a tiny silicon mass.
Intel is advising customers to simply NOT overclock their CPU ...
If it IS the case that it's just physics, then Intel has hit a wall re: their enthusiast customer base -- at some power magnitude you just can't get the heat out well enough to maintain a live-able temp. I wouldn't expect an Intel engineer to state that; I'd expect to hear a marketing type say 'just don't overclock'.
Maybe I'm being tremendously obvious, but this is my opinion re: the tons of posts I'm seeing about the 7700K being somehow flawed and that's why the usual OC routines aren't successful. Maybe there IS a design flaw, but it could just be physics.
