Info 64MB V-Cache on 5XXX Zen3 Average +15% in Games

[Kedas]

Well we know now how they will bridge the long wait to Zen4 on AM5 Q4 2022.
Production start for V-cache is end this year so too early for Zen4 so this is certainly coming to AM4.
+15% Lisa said is "like an entire architectural generation"

 

[MadRat]

I've had dead CPUs snap to life after a hard bios reset. Those old Durons were funny like that.

 

[Schmide]

I'll do the full rundown.

replaced 5800x with 5800x3d, system had been running for half a year+
booted asked to clear keys say yes
go to bios set xmp profile to settings 3466 works well for my 64gb 4x sticks of 3600
do quick fan setup
reboot return to bios check settings
reboot go to windows. pump and fan settings reversed
reboot to bios fix fan settings. reboot. (never comes back)
clear bios
nothing

further troubleshooting
4 sticks of ram to 2 sticks to 1 sticks to 1 stick of other ram i have 3200
swap gpu 6900x for 3070ti
check all power.
put 5800x back in. works fine.

put 5800x3d into gigabyte aorus 570x (had 3900x)
reset bios nothing
put 3900x back in works fine

It's dead. Even if there was a way to bring it back from the dead it's a lemon. There is enough working hardware to confirm the 5800x3d is the bad unit.

 

[Dave3000]

Have you updated to the latest BIOS?

 

[Schmide]

Yes on both motherboards

 

[DrMrLordX]

At least when I killed my first 3900X, I knew exactly how and why it died. Suicidal CPUs are confusing and just suck in general. At least you get an RMA.

 

[BTRY B 529th FA BN]

@Schmide did you buy it from a Brick and Mortar store or online? both mine have been from Micro Center in-store bought.

 

[Schmide]

Too late for that. It will have to run its course through AMD

 

[MadRat]

When I think of hard reset I revert back in BIOS/system level, too. Had the same behavior with Durons. Thought they were dead but booted fine on unsupported Socket A and would begin working again on newer systems. Almost like they had memories.

 

[DAPUNISHER]

Booted, changed xmp profile, changed fan profile, got to windows, rebooted, change fan profile again. got D0 code. Tried with different sets of memory and video card. Tried another motherboard. It's dead jim

If you were letting the board handle voltage and boost with its default profile, I suggest taking control of it this time around.

 

[Timorous]

Installed it in my B350M Mortar tonight. Did some testing first CB R23 run I got 14k points but I did load up reddit part way through to look at temps because it was hitting 82 degrees.

Tried PBO2 Tuner and -20 on all works so I will just leave it there. Temps did not drop when re-testing CB R23 but I got 15k points instead and clocks were a lot higher so I will take that.

Botted up Path of Exile to check the new league and see what it was like. CPU frame time in the 1st town is 2ms which is a massive improvement from my 2200G that was closer to 10 and would cause frame rate drops. Will have to do more testing on that but don't have time now. Temps in Path of Exile are around 66 degrees which is not too bad.

 

[Schmide]

If you were letting the board handle voltage and boost with its default profile, I suggest taking control of it this time around.

Memory voltage for 3466 is 1.356. No way it would/should fry a chip. Especially not in 5min.

 

[Markfw]

Memory voltage for 3466 is 1.356. No way it would/should fry a chip. Especially not in 5min.

Could have been the CPU voltage that was going nuts.

 

[Schmide]

Could have been the CPU voltage that was going nuts.

I checked the temps when I booted to windows, max temp on all the cores was like 42c and it's cooled by an EK waterblock. Any chip that gives up the ghost with such little usage is faulty.

 

[jpiniero]

I checked the temps when I booted to windows, max temp on all the cores was like 42c and it's cooled by an EK waterblock. Any chip that gives up the ghost with such little usage is faulty.

I'm betting you still killed it somehow. Water sounds like overkill for 5800X3D.. did you turn PBO on?

 

[gdansk]

Installed it in my B350M Mortar tonight. Did some testing first CB R23 run I got 14k points but I did load up reddit part way through to look at temps because it was hitting 82 degrees.

Tried PBO2 Tuner and -20 on all works so I will just leave it there. Temps did not drop when re-testing CB R23 but I got 15k points instead and clocks were a lot higher so I will take that.

Botted up Path of Exile to check the new league and see what it was like. CPU frame time in the 1st town is 2ms which is a massive improvement from my 2200G that was closer to 10 and would cause frame rate drops. Will have to do more testing on that but don't have time now. Temps in Path of Exile are around 66 degrees which is not too bad.

Which BIOS version are you using? Out of curiosity

 

[Schmide]

I'm betting you still killed it somehow. Water sounds like overkill for 5800X3D.. did you turn PBO on?

Honestly this is like slut shaming. I explained every step very clearly and I have been building systems since the 80s. I've killed my share of hardware but in no uncertain way did I expose this chip to anything but the best environment.

Edit: and btw the main reason I water cool is for silence.

 

[Ranulf]

I'm betting you still killed it somehow. Water sounds like overkill for 5800X3D.. did you turn PBO on?

Uh, okay. I thought the 5800x's known higher temps was a reason to put it under water. And I say this as someone who much prefers air cooling.

 

[BTRY B 529th FA BN]

These 3D chips are real sensitive to vsoc volts. Before on my 5800X I needed a little over 1.1v for 1866, with this new production finer built chip it can do 1900 1v so there's been some changes. I was pushing 1.0875vsoc and was getting some weird instability until I lowered it.

 

[DrMrLordX]

Honestly this is like slut shaming.

Hey you be slutty all you want! There's no shame in it.

As for the CPU, my guess is that some transient voltage spike might have knocked out the 3D cache die. It can't handle volts as high as the CPU, which is why they tried putting a hard upper limit on the amount of vcore the chip can draw (no extra voltage rail means the 3D cache die is running at the same voltage as the CPU). Some boards misreport power/voltage to the CPU so it may be difficult to prevent 5800X3D chips from going over 1.35v in every situation.

These 3D chips are real sensitive to vsoc volts. Before on my 5800X I needed a little over 1.1v for 1866, with this new production finer built chip it can do 1900 1v so there's been some changes. I was pushing 1.0875vsoc and was getting some weird instability until I lowered it.

That may be more a feature of the B2 stepping than the extra cache die.

 

[Timorous]

Which BIOS version are you using? Out of curiosity

Latest beta that is agesa 1.2.0.7. They updated it a few days ago but only just got around to installing it.

 

[Timorous]

These 3D chips are real sensitive to vsoc volts. Before on my 5800X I needed a little over 1.1v for 1866, with this new production finer built chip it can do 1900 1v so there's been some changes. I was pushing 1.0875vsoc and was getting some weird instability until I lowered it.

I will have to try memory tweaking. I set my ram to its 3600C16 XMP profile, set fclk to 1800 and soc voltage to auto and it seems to be working fine so far. Not sure if I should push for higher speed at c16 or see if I can improve latency. My 2200G always failed to boot at 3600C16 so that was also an instant improvement.

Spent most of last night playing with mounting and fan profiles because to begin with temps were really high. Took my case apart and took the fans off the radiator to find it caked in dust so cleaning that dropped temps by 5 degrees. Still hot but it is no longer above 85 when running CB r23 so I can live with that.

Might also do more PBO2 tuning, set it to -20 all cores and it did an r23 run no problem, didn't really drop temps much but all core clocks were higher. Since -20 worked fine I will try lower settings to see what it can do.

 

[Schmide]

Hey you be slutty all you want! There's no shame in it.

I don't think you understand the term. If you want me to explain it I will but otherwise it is not something to celebrate. No user should go through unfounded accusations that they did something wrong when they didn't.

As for the CPU, my guess is that some transient voltage spike might have knocked out the 3D cache die. It can't handle volts as high as the CPU, which is why they tried putting a hard upper limit on the amount of vcore the chip can draw (no extra voltage rail means the 3D cache die is running at the same voltage as the CPU). Some boards misreport power/voltage to the CPU so it may be difficult to prevent 5800X3D chips from going over 1.35v in every situation.

You're searching for blame in the motherboard. It died in 5min and nothing was misconfigured. The chip should be able to tolerate some fluctuation in voltage. It was faulty. It happens sometimes. If its replacement dies that's another story. A trend of 1 means nothing.

 

[blckgrffn]

I don't think you understand the term. If you want me to explain it I will but otherwise it is not something to celebrate. No user should go through unfounded accusations that they did something wrong when they didn't.



You're searching for blame in the motherboard. It died in 5min and nothing was misconfigured. The chip should be able to tolerate some fluctuation in voltage. It was faulty. It happens sometimes. If its replacement dies that's another story. A trend of 1 means nothing.

Let me be clear - I appreciate you sharing your experience here. I am also disappointed that the first and ongoing reaction has been accusations of “doing it/something/anything wrong”. I too would come here to commiserate the loss of a new CPU if it happened.

I have two 5800x3D chips to use in builds now and will be keeping an eye on them - I am building with a 3400G to boot/confirm other hardware and update firmware. Whatever my experience I too will share it.

Stuff happens, but if those of us using the chip have similar experiences that would be troubling. If it turns out to be a super anecdotal experience that’s good for us.

 

[DrMrLordX]

I don't think you understand the term.

It's called a joke. I know exactly what it is.

You're searching for blame in the motherboard.

IMPORTANT - Explaining the AMD Ryzen "Power Reporting Deviation" -metric in HWiNFO

Ryzen CPUs for AM4 platform rely on external, motherboard sourced telemetry to determine their power consumption. The voltage, current and power telemetry is provided to the processor by the motherboard VRM controller through the AMD SVI2 interface. This information is consumed by the processors...

www.hwinfo.com

In case you need a reminder, many AM4 boards are guilty of feeding incorrect power levels to CPUs. The 5800X3D is the first CPU that has a firmware vcore lock to protect key components. There must be a good reason why AMD won't even let "xtreme" overclockers give extra vcore to the chip. Nobody seems to know yet what , if any interactions these power deviation "cheats" may have with the 5800X3D. In case I have to spell it out for you, IT ISN'T YOUR FAULT YOU DIDN'T DO ANYTHING WRONG.

 

[igor_kavinski]

Some of the gains even at 1440P are crazy.