Unable to achieve 5GHz speed (last 4.9GHz) at 4790k
Unable to achieve 5GHz speed (last 4.9GHz) at 4790k
Understanding what vDroop entails and how to manage it are distinct challenges. Today, most users don’t need to intervene much; the system manages this internally.
Here’s a summary of vDroop:
This concept was initially more relevant to older hardware. Its purpose is to avoid over-voltage conditions that could harm or damage the CPU. It ensures that the voltage level set in the BIOS doesn’t surpass safe limits.
LLC aims to maintain stable voltage levels, so its two functions operate in opposing directions. On Intel platforms, a small voltage difference—around 0.005 to 0.01 volts—can significantly impact clock speed performance. This requires precise adjustments. The effectiveness largely depends on the quality of components on the board, which is often beyond customer control.
Setting LLC to a higher value is like releasing a dam’s gate, allowing excessive voltage to flow. Comparing it to a hydroelectric facility, high LLC with substantial voltage could lead to severe consequences for the chip.
The only reliable method to manage vDroop is through continuous voltage monitoring. Start by resetting to defaults and disabling LLC. Compare the reported voltage from your monitor against the one set in the BIOS. If they match, you’re likely within safe limits; if not, further investigation is needed.
At default settings, the system automatically applies the standard voltage, which should align with what you see on the monitor. To test for vDroop, manually adjust the voltage and observe stability. If it remains consistent with your target, proceed confidently.
If no vDroop appears at default levels, ensure power-saving states are disabled as mentioned. As voltage increases, incrementally check each stage in small steps—typically 0.1 GHz and 0.01 volts—to identify the optimal setting. If the voltage drops below target, adjust LLC accordingly, but always stay within recommended limits. The board’s components set this boundary; exceeding it risks damage.
Monitoring temperature alongside voltage is crucial for overclocking. Fine-tuning is complex and demands significant expertise. Managing heat is more straightforward.
Custom cooling solutions can be costly and risky. Investing in a high-quality cooler isn’t always necessary, especially if you’re not willing to take the risk of damaging your CPU. The decision hinges on balancing performance gains against potential hazards.
Understanding these factors helps avoid common pitfalls. Adjustments should reflect deeper knowledge than what the BIOS provides, and excessive experimentation can be detrimental.
You also have to take into consideration exactly what you are looking at as far as vcore voltage reporting goes. There's a rather large difference between VID and core voltage and software can use either and often does. For instance for me, in cpu-z ROG edition, that says vcore but in cpu-z MSI edition it says VID, and those voltages will be different. VID is what the cpu is asking the motherboard VRM's to supply (upto specific max value), vcore is what the cpu is actually using, independent of whatever 'vcore' setting you apply. VID is accurate as it's a cpu reported value, vcore is software read values, so can be 20%± inaccurate.
Take software that says 'cpu voltage' or similar with a grain of salt, there's no telling if it's vcore or VID or something else entirely (amd fx was bad for that)
Maximim safe value for 4th gen Intel is 1.4v, however, you realistically want to be as far under that value as possible because of Vdroop, VRM 'theft', inaccuracies in vcore reading. 1.316v is the highest I'd safely agree to, 1.45v is asking for burned out transistors in your cores. Too much of that and bye bye core, bye bye cpu.
That psu is 80+. That means it's supposedly been voluntarily submitted for testing and gets @80/80/80% at 20/50/100% loads. Not 85% or that'd be Bronze, not White. It has 4 separate rails, not 4 combined rails like better units, cpu/pcie/mobo etc and is a really old group regulated design carrying such high amperage on the 3.3/5v+ rails. I'd not try any OC with that psu, in simple terms it's a fireworks show waiting to happen.
TXM Gold is the better platform vs the FSP G, but both are decent for high OC stability, which you will not get with your current psu, far too erratic ripple that's probably far out of ATX specs.
Thank you for your quick response. I will keep the system within 1.316v @4.9Ghz and avoid going above 5Ghz at this time.
Thanks for answering. If there is a matter of 80% then it means that there is fighting start between components to draw power because what I understand 575W combine power x 80% = 460Watts. Damnnnnnnn you are right and I understand this very well that when a good component start putting extra efficiency to get proper power to itself then this thing weaker its strength. We can understand this thing as if a human can carry weight daily, where he/she has to put extra strength / power than usual than soon he/she become weak.
So thank you very much. I will first go with
Corsair TX750M 750W 80 Plus Gold Semi-Modular Power Supply
If not available then I will go with FSP.
Thank you once again for your time, help and guidance. really appreciated much much much...... Thank you again bro for your help and guidance.
Maybe it would be better to stick with a running speed of 4.8 instead of going deeper. Their advice about a vCore of 1.35 or 1.4 was not well received. The only improvement you could make is looking for more trustworthy sources. I believe the solution would simply be setting the vCore to 1.3, which might help. Stronger cooling could also assist in reaching the higher clock speed, though it would require a significant investment for the custom loop.
Thank you for your response.
1) Vcore
I prefer not to run a continuous processor at 24/7 (4–5 hours daily usage and gaming). A 4.9Ghz processor with 1.316v is not ideal, especially when using auto voltage (HWiNfo 1.377v during Cinebench). The maximum temperature reached was 67°C.
The Intel Burn Test passed. Your suggestion of 4.8Ghz at 1.275v for daily use and gaming—even with heavy CPU-intensive games—seems very stable, keeping temperatures below 60°C.
Thanks to the power supply; I’m now back to 4.6Ghz with 1.225v under 40°C, which is within safe limits (especially during cooler weather).
Which option would be best if you have a reliable 750W power supply?
2) Power Supply
You mentioned the PSU isn’t ideal, meaning the processor isn’t performing well.
Anyway, I’m now considering these options at the same price: US$ 91.42.
- Thermaltake Toughpower Grand RGB 750W Gold Fully Modular
- FSP Hydro G HG750 750W 80 PLUS GOLD Certified Full Modular Active PFC Power Supply
In a review, the Hydro G HG750 was noted to have a noisy fan, which is more bothersome than the radiator fans of the H100i V2.
I’m unsure if FSP has made adjustments for quieter operation. Sorry for repeated concerns.
I’m not sure about the Corsair TX750M, but I feel uncertain about it. It might be a sign that I should stick with the FSP model I’ve been using for the past few years.
I haven't spent a lot of time examining power supplies, but to be cautious, it's a good idea to get a replacement. I didn't see it as a problem when purchasing my FX and my system likely isn't very demanding. It could become necessary after 4-5 years, but it shouldn't prevent higher overclocking. It's possible that no matter what you do, the chip just won't perform well.
Let me know if you want a more detailed explanation or further clarification.
Because it's a 4 or 5 year old model, it shouldn't cause any problems. The documentation mentions it could last 5 to 10 years. Usually people replace their PCs before the power supply becomes outdated, so the rules have shifted a bit. So far, PSUs haven't been an issue for me, but I haven't used high or mid-range systems. One day you might notice it won't turn on, which depends on whether you're planning an upgrade soon. It seems like it still has some capacity left.