What are the risks of adjusting the BCLK value from 99.76 to 101 on a laptop?
What are the risks of adjusting the BCLK value from 99.76 to 101 on a laptop?
Laptop model details: Asus ROG Strix G614
CPU specifications: i7-13650HX
I'm exploring overclocking and am new to the process. The default bus speed is 99.76, and I'm trying to set it to a whole number for stability at 52 multipliers. However, it currently caps at 51. When I adjust it to 100 in xtu, it functions normally at the default 99.76. Will setting it to 101 cause any issues? Please let me know if you need further clarification.
Intel typically applies a 0.5% down spread on BCLK by default. This affects how values are rounded—when you set 100, it's adjusted from 99.5 to 100, resulting in an average of about 99.76. Using center spread or turning it off usually yields a cleaner 100MHz result. However, Intel generally restricts center or up spread settings. If you're not getting a precise 100MHz, consider raising BCLK to 100.25MHz with down spread, though the PLL's resolution may be limited and could introduce slight discrepancies.
Most likely. Since at 101 BCLK, CPU wouldn't be stable.
Also, there is 0 gain in performance if you increase the BCLK.
99.76 BCLK with 51 ratio = 5087 Mhz or 5.087 Ghz
99.76 BCLK with 52 ratio = 5187 Mhz or 5.187 Ghz
100 BCLK with 51 ratio = 5100 Mhz or 5.1 Ghz
101 BCLK with 51 ratio = 5151 Mhz or 5.151 Ghz
100 BCLK with 52 ratio = 5200 Mhz or 5.2 Ghz
101 BCLK with 52 ratio = 5252 Mhz or 5.252 Ghz
When your CPU OC is stable with 99.76 BCLK at 52 ratio, for 100 BCLK at 52 ratio, you'd only gain 13 Mhz (5.187 Ghz vs 5.2 Ghz).
Even a gain of 200 Mhz would be peanuts and not worth the effort.
Back in the day, with older CPUs, CPU OC was worthwhile.
E.g i have i5-6600K with 3.5 GHz base and 3.9 Ghz boost. With CPU OC, i could get it 4.5 Ghz all core (increase of 600 Mhz over boost), or with delid, ~4.7 Ghz all core (800 Mhz over boost). And there have been some delidded i5-6600K CPUs, that can hold 5 Ghz all core.
Essentially from Intel 12th gen and onwards, most chips out there can only hold all core stable 100-300 Mhz over max boost. That gain is so little, that CPU OC with current, highly efficient chips, isn't worthwhile. There won't be any meaningful performance increase.
If the headroom would be bigger, like it is with my 6th gen CPU, where on minimum, i look towards 600 Mhz increase over boost clocks (or up to 1.1 Ghz over boost, if very lucky with delidded chip), then CPU OC makes sense.
All-in-all, CPU OC is dying niche and outside of record breaking, isn't worth the effort anymore. Better to run stock clocks and let CPU to decide when to turbo up. Less energy waste and less heat production this way also. Not to mention CPU lifespan, since when running stock clocks, CPU lifespan is easy 10+ years. Running all core OC 24/7 will reduce CPU lifespan considerably. E.g if i were to run 4.5 Ghz on my i5-6600K, i could cut the CPU lifespan in half. And when running CPU at high OC levels, the absolute maximum CPU is able to run at (e.g ~4.7 Ghz on my i5-6600K), the chip burns out in 1-2 years.
XTU has it's limitations and it can be iffy when showing CPU stats.
Using XTU is like slapping a turbo to an engine and hoping to get the best results, without doing the chip tune and dyno, for optimal performance.
Intel XTU, MSI OC Genie, AsRock OC Tuner, Asus AI Overclocking, Gigabyte EasyTune etc, are all lazy man's OC options.
With essentially 1 click (hence why "lazy man's OC"), they put some level of OC on CPU (or whole system), while going way too high with voltages, among other things. Moreover, while it may give a bit better performance, those "profiles" won't tell what changes they do within BIOS. This can result in all kinds of stability issues. Even hardware failure.
All-in-all, if one wants to OC their CPU, better do it manually from BIOS. Or not do anything at all. This "convenient" 1 button/click OC is bad for hardware. Always has been.
Source: Intel -
https://www.intel.ca/content/www/ca/en/g...cking.html
Give it a read, it's a good read when starting to learn about CPU OC.
Since you have a laptop, do note that you will hit thermal ceiling, fast.
Now, if you would have desktop, where you can go big on CPU cooler, thus increasing thermal ceiling, CPU OC would make a more sense.
If I were you, I would keep the BCLK value unchanged because it affects the rest of your platform's I/O. The 13th Gen Intel platform for both desktop and laptop has already failed due to a manufacturing issue, so adding another reason for instability wouldn't help. Instead, I'd lower the processor and GPU voltage, replace the thermal pads with higher-quality ones, and then stop further changes.
Here’s the revised version maintaining the original meaning and structure:
I appreciate this useful information. To sum up... In Cinebench R23 with multi-core, I haven’t encountered any cores surpassing 81°C at a steady 5.11GHz. I’m quite pleased. Only once during the Aida 64 test did I reach 84°C, but my core clock remained stable. During the OCCT AVX2 test, it operated at 5.02GHz for the first minute and 15 seconds; once the temperature hit 95°C, it dropped to between 4.80 and 4.90°C and stayed there permanently.
I would have preferred the system to maintain a consistent 5GHz frequency, but I couldn’t push further in any settings. Several reboots occurred afterward, particularly when the charger wasn’t connected. So, I had to lower the voltage while keeping the core multiplier steady at these highest levels. HT and ECORE were disabled, by the way. Overall, I’m enjoying these configurations. The transition from a peak of 4.20GHz to 5.10GHz was satisfying. I also have a cache offset setting of -0.060, which I didn’t capture in a photo. Thanks for your assistance—I’d love to hear your thoughts.
I mistakenly shared the incorrect profile picture; core offset is -220 and healthy employee.
imgur.com/a/gTkWHNZ
By the way, the temperature in my city is around 34°C. I’m unsure how much warmer it gets inside the house. I also set the fans to full speed when temperatures exceed 70 degrees. There are three fans—CPU and system fans are at maximum (CPU 6800rpm, system fan 7600rpm, though slightly smaller). Once the temperature reached 70°C, my GPU fan started spinning at 3200rpm.
It’s currently 2:04 pm here.
Yes, I agree, but I'm interested in finding out if a 1 MHz rise could lead to adverse effects. I don't want any extra issues. Thanks for your assistance.
Friends, I don't speak English well. Please forgive any mistakes in the translation. I'm ready to address any questions you have.
Not my own, yet sometimes I feel more comfortable with English than my original language.
I question why so many original characters aim for extremely high CPU speeds—beyond personal preference?
Maintaining such high CPU frequencies continuously is similar to keeping your engine's RPMs at 8000 constantly. Whether you're driving in the city, on the highway, or idling with the engine running, it never drops. Would you do the same for your car? If not, why would you?
Running the CPU at its maximum will quickly degrade its lifespan.
Besides, it uses around 157W continuously (which is probably cheap electricity), and at those high frequencies it generates a significant amount of heat.
And all that... just for what?
Or are you curious about how long your CPU lasts before an overclock causes damage?
It’s more about the tiny gains—like moving from 4.9 GHz to 5.1 GHz, or 200 MHz extra.
That 4.9 GHz is the upper limit for your CPU turbo.
Specifications:
https://www.intel.com/content/www/u...-2...tions.html
Is 1 MHz or 1 GHz better?
To boost CPU speed by just 1 MHz, you only need to raise the BCLK ratio by 0.01. That gives an extra 1 MHz.
Even with your best efforts, reaching 100 MHz might be impossible.
Clock signals come from physical parts (oscillators) that naturally vary during manufacturing. These variations can slightly shift the actual frequency from the planned 100 MHz.
Moreover, manufacturers may opt for a lower BCLK to enhance stability or work better with other components. Lower frequencies can reduce signal integrity risks and improve reliability.
Some systems use spread spectrum clocking, which intentionally fluctuates the clock signal to minimize electromagnetic interference (EMI). This fluctuation can also make the BCLK appear slightly different from 100 MHz.
A minor deviation—like 99.76 MHz instead of 100 MHz—doesn’t affect performance much. The CPU can adjust its multiplier and other parameters to handle the slight change.
It was difficult to detect 4.9GHz on this processor. Possibly one or two cores, perhaps. Currently I can run all cores at 5.1GHz, maybe even higher, but the average effective clock speed is 5.1GHz, that's what I see. I intended to lock it at 5GHz in AVX2. Of course, 1.6GHz at idle. I'm using a balanced power profile, by the way.
4.9 Ghz P-core turbo can indicate either a single core or occasionally multiple cores. Typically, only one core achieves 4.9 Ghz, while the others cap at lower frequencies (4.8 Ghz, 4.7 Ghz). However, not every sixth core will hit 4.9 Ghz unless you manually adjust the chip. The simplest setup is a BCLK of 100 Mhz with a 50 multiplier. But this BCLK might lack stability. For a BCLK of 99.76 Mhz, the multiplier should be around 50.12 (approximately 4.999 Ghz) or 50.13 (approximately 5.001 Ghz). At 98.04 Mhz BCLK with a 51 multiplier, you reach about 5.000 Ghz.