AMD Announces BIOS Fix for Ryzen 3000 Boost Clock Issue

Tsing

The FPS Review
Staff member
Joined
May 6, 2019
Messages
11,255
Points
83
AMD has finally admitted that the boost clocks of its third-gen Ryzen processors have failed to meet advertised specifications. The company is making amends with a future BIOS update, which should provide the level of performance that was originally promised to enthusiasts.

AMD is blaming the issue on a firmware problem that reduced boost frequency "in some situations." It isn't clear when the fix will be available, but the company will have more to share on September 10.

While processor boost frequency is dependent on many variables including workload, system design, and cooling solution, we have closely reviewed the feedback from our customers and have identified an issue in our firmware that reduces boost frequency in some situations.
 
I don't think "some situations" is gonna be sufficient to bail them out of this one.
 
When they say "some situations" what they really mean is "all situations".
 
If you're running on sub standard cooling, with **** for voltage, and expecting your first gen cheapest you could find Ryzen motherboard to run the new chips at max boost simply because you managed to shoehorn in a new BIOS... think again.
 
Grimlakin said:
If you're running on sub standard cooling, with **** for voltage, and expecting your first gen cheapest you could find Ryzen motherboard to run the new chips at max boost simply because you managed to shoehorn in a new BIOS... think again.
Click to expand...
I wouldn't consider the Asus Crosshair VII (X470) motherboard to be a "Cheap" motherboard, and I've got an AIO 240mm Radiator for cooling my 3700X. I'm not even remotely CLOSE to being able to get advertised boost clocks, even on a single core..
 
What is your power supply? And no I wouldn't consider that bargain basement and would expect the update to help unleash the full potential of your CPU. If it doesn't I would be surprised.
 
See my sig. X570 Aorus Ultra w/ a Season Prime Ultra 1000. There's not a cheap component on my build.
 
Grimlakin said:
What is your power supply? And no I wouldn't consider that bargain basement and would expect the update to help unleash the full potential of your CPU. If it doesn't I would be surprised.
Click to expand...

Seasonic FOCUS Plus Series SSR-750PX 750W 80+ Platinum
 
I'm imagining that most people on the enthusiast forums don't fall into the bottom rung of budget builders. But when I was a young man and computer enthusiast I sure as **** did. I remember my Athalon 3200+ build I did with Windows vista. It was scraping the cheapest but branded parts I could get.
 
Riccochet said:
When they say "some situations" what they really mean is "all situations".
Click to expand...

Well "some situations" could be incredibly common without being all situations by any definition of all.
 
My gut is telling me this may have more to do with unstable power than quality components. For those of you not seeing max auto over clocking on a core, and actually more importantly perhaps, for those of you that ARE seeing it.

1. Are you on a UPS.

2. Does that UPS do power or line conditioning to remove variance and spikes from the power?

Personally I got a cyberpower UPS a few years ago and it's given me amazingly clean power ever since. Regardless of what's going on in the rest of my grid/home. I'm not running AMD but I remember back when the IBM Deskstar (aka death star) 80 gig drives were failing left and right and it was because they were enterprise class drives being put into consumer hardware. Consumer hardware that didn't have clean stable power and the drives couldn't handle it.
 
David_Schroth said:
A good power supply should also buff out all but the dirtiest of power. See: PSU reviews that Paul does that hits them with 100v input voltage.
Click to expand...

Even a good power supply won't buff out bad fluctuations in incoming power.
 
Riccochet said:
When they say "some situations" what they really mean is "all situations".
Click to expand...


Nah. More like 85% of situations, judging by reviewers results.

A minority of testers were actually able to get full advertised boost clocks. There weren't many of them though.
 
There seems to be no rhyme or reason to it besides silicon lottery. I can take the 3900X and put it on the ASUS Crosshair VIII Hero, which won't get close to the advertised boost clocks. I can take that same CPU and put it on the MSI MEG X570 GODLIKe, and get 4.575GHz. I can change the RAM and either stop it from hitting the correct boost clocks or make it get close again. On the same board the 3900X doesn't boost close to the correct clocks, the 3700X will. As I said in my 3700X review, I tried our sample CPU which does boost nearly at the correct speeds and a retail CPU which doesn't get all that close for the most part.

In other words, I can change RAM, motherboard, or CPU and it may make a difference or it might not.
 
Dan_D said:
There seems to be no rhyme or reason to it besides silicon lottery. I can take the 3900X and put it on the ASUS Crosshair VIII Hero, which won't get close to the advertised boost clocks. I can take that same CPU and put it on the MSI MEG X570 GODLIKe, and get 4.575GHz. I can change the RAM and either stop it from hitting the correct boost clocks or make it get close again. On the same board the 3900X doesn't boost close to the correct clocks, the 3700X will. As I said in my 3700X review, I tried our sample CPU which does boost nearly at the correct speeds and a retail CPU which doesn't get all that close for the most part.

In other words, I can change RAM, motherboard, or CPU and it may make a difference or it might not.
Click to expand...

hummm that suggests there is a lot to this Bios update coming.
 
You must log in or register to reply here.
Become a Patron!
Back
Top