Never seen boost clock above 5650mhz on a 7950x... (pbo, CO, eClk, all tried)

[nirurin]

I've tried core-optimiser, I've tried eClk, I've tried +fmax, I've tried 1X, 4X, 8X and 10X scalars...

No matter what I do, the max -effective- clock in hwinfo is always 5650 on my best core.

With all those things added in, it's currently 'requesting' 5975mhz, instead of the previous 5800, but thats irrelevant as the effective clock always seems to be 5650.

Which bugs me lol, cos I've seen a lot of people with at least 5700, if not 5800, using similar cooling to me.

Currently on arctic liquid freezer 2, and my temps even under 230watts never hits 95C, usually more like 85C, so I can't see cooling being the main issue...
(Single core temps only ever get to about 70, if that).

[Soupladel]

nirurin wrote:
I've tried core-optimiser, I've tried eClk, I've tried +fmax, I've tried 1X, 4X, 8X and 10X scalars...

No matter what I do, the max -effective- clock in hwinfo is always 5650 on my best core.

With all those things added in, it's currently 'requesting' 5975mhz, instead of the previous 5800, but thats irrelevant as the effective clock always seems to be 5650.

Which bugs me lol, cos I've seen a lot of people with at least 5700, if not 5800, using similar cooling to me.

Currently on arctic liquid freezer 2, and my temps even under 230watts never hits 95C, usually more like 85C, so I can't see cooling being the main issue...
(Single core temps only ever get to about 70, if that).

You have not said which board you have and what version of the BIOS you are using.

I am having what appears on the face of it to be the exact same problem, but in my case, all but one of my course are limited to 5.5Ghz boost, with core 0 boosting very occaisonally to 5.765.

I am using the Crosshair X670E Hero motherboard and all BIOS versions after 0705 have the same boosting behaviour, which i have been led to believe is caused by the latest AGESA patch used in the newer versions of my BIOS.

[nirurin]

Soupladel wrote:
You have not said which board you have and what version of the BIOS you are using.

I am having what appears on the face of it to be the exact same problem, but in my case, all but one of my course are limited to 5.5Ghz boost, with core 0 boosting very occaisonally to 5.765.

I am using the Crosshair X670E Hero motherboard and all BIOS versions after 0705 have the same boosting behaviour, which i have been led to believe is caused by the latest AGESA patch used in the newer versions of my BIOS.

Haha, I just came back to the forum, because I had realised this exact thing. I forgot to mention my setup.

Rog Strix e-e, using the 803 test bios, but I was on the latest 'official' bios before that and it seemed to be the same. (So its Agesa1.0.0.3 patch A in both cases I believe?)

[Tyberious]

nirurin wrote:
Haha, I just came back to the forum, because I had realised this exact thing. I forgot to mention my setup.

Rog Strix e-e, using the 803 test bios, but I was on the latest 'official' bios before that and it seemed to be the same. (So its Agesa1.0.0.3 patch A in both cases I believe?)

0803 is actually Agesa 1.0.0.3 Patch C

[Soupladel]

nirurin wrote:
Haha, I just came back to the forum, because I had realised this exact thing. I forgot to mention my setup.

Rog Strix e-e, using the 803 test bios, but I was on the latest 'official' bios before that and it seemed to be the same. (So its Agesa1.0.0.3 patch A in both cases I believe?)

Yes, I believe everything after 1.0.0.3 patch A, so far, has the same boosting behaviour limitationsÂ*

[Shamino]

theres no way effective clk = running clk unless the thread/core is busy 100% of the time.

[nirurin]

Shamino wrote:
theres no way effective clk = running clk unless the thread/core is busy 100% of the time.

I know this, but many people specifically say their 'effective clocks' are hitting 5.8+ mhz. Or even 5.7+. Mine seem stuck at 5650 on my best core.

[Shamino]

then they are doing some work on their core when taking the screenshot

[nirurin]

Shamino wrote:
then they are doing some work on their core when taking the screenshot

I'm not sure what you're getting at... I'm also doing 'work' on the cores when testing. Thats how I know the boosts. If I wasn't doing work, they'd be sat at under 100mhz parked.

Also the Asus boards definitely have some weird power bug. I can push 250W into the PPT of the chip, but the cpu temps only get to about 85C, and I barely get 38500 in cinebench.
On an MSI board I can get 38500 using 200watts, and the same temps. So Asus is taking 50w and just... throwing it away somewhere.