• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Ryzen 7 9800X3D

Grim5 said:
Tech Jesus is quick on the case, to investigate why this 9800x3d blew up








Click to expand...
I wouldn't worry about it personally, the images the person uploaded you can see there is physical damage at the top of the socket which would have likely pushed the chip slightly off so contact short.

It's not known if that was shipped to them damaged or they did it via user error, id be shocked if it was the exact same issue as last time as this was added to AMDs validation process for boards partners.
 
Solaris said:
So my chip arrived at 10:30am this morning.
DPD guy left it on my doorstep with no knock which is surprising, but I got my delivery e-mail and went to retrieve it.

Shut down my PC, whipped the AIO straight off the old CPU while it was still warm and before the TIM had a chance to cool down and solidify.

Straight in with the screwdriver, whipped out the old motherboard and out comes a standoff, annoying.

So I go and find a tool that will remove the standoff, screwdriver on one side, deep hex socket on the other, comes off fine.

I turn around to speak to my missus, the standoff still in the hex socket (I think...), conversation over I go to look for the standoff... Gone... vanished...

I spend the next hour searching the floor, looking through various piles of clutter, can't find the blasted thing anywhere.
I go through my spare screws and fittings looking for something suitable, nothing....
I end up going to a spare Lian Li case I have in the cupboard, remove a standoff from it, and use that.

Panic over.

5800X3D goes in new motherboard.
New motherboard goes in.
Cable everything up.
First boot.

No video output, and some absolutely garbage error codes on the onboard debug display.

Maybe the BIOS is incompatible with such a new CPU.

Grabs a USB stick, uses a laptop to format it to FAT32 and whack the latest BIOS on it, ranaming it to MSI.ROM.

Seems to flash okay, but still nada.

The next two hours are spent checking every cable, reseating everything, trying different combinations of things plugged in and unplugged.

I was contemplating RMAing things, but was it the board, or the RAM, or the CPU?? I couldn't figure it out.

Suddenly with all the USB headers unplugged and only 1 stick of RAM in I get the first good boot to BIOS.
I swap RAM sticks, no boot... okay so it must be 1 bad RAM stick.
I swap back, still no boot... so it must be the motherboard or memory controller on the CPU?

Then it dawns on me. The one variable I haven't tried changing yet is a Samsung 990 Evo PCI-E Gen5 SSD in the first NVME slot. I hadn't thought about it as it was hidden under a metal heatsink.

I tear it out, put a Western Digital PCI-E Gen4 one in the slot. Put the heatsink back on.

And it lived.

All the RAM was good, the CPU was perfect, I could reconnect all the headers and drives and cables.

I even put the troublesome SSD in a different slot and it worked fine.

Panic over. I think I still have some hair left...

(Turns out the garbage error codes are a temperature readout that sits on the EZ Debug screen after a "good" boot, not helpful MSI, and not even mentioned in the manual!?!?! I was wondering why it was bouncing between error codes 40, 41, and 42 which aren't error codes mentioned in the manual... It's CPU Idle temperature!!)
Click to expand...
These sort of things seemed to be, for the most part, a thing of the past, until the last few years. I remember about 20 years ago, you'd always have some sort of issue like the above to troubleshoot with new hardware.

I keep seeing a lot of this sort of thing, especially with the AMD stuff, although it hasn't put me off upgrading to the 9800 from an older intel imminently.

Or is it just me being out of touch a little?
 
D-Ranged said:
These sort of things seemed to be, for the most part, a thing of the past, until the last few years. I remember about 20 years ago, you'd always have some sort of issue like the above to troubleshoot with new hardware.

I keep seeing a lot of this sort of thing, especially with the AMD stuff, although it hasn't put me off upgrading to the 9800 from an older intel imminently.

Or is it just me being out of touch a little?
Click to expand...
It's always going to be a thing in computer hardware.
I'm guessing the PCI-E Gen5 NVMe was stealing PCI-E lanes from the GPU that it needed but the rest of the PC was booting fine.

Part of the fun tbh.
 
Hicks12 said:
I wouldn't worry about it personally, the images the person uploaded you can see there is physical damage at the top of the socket which would have likely pushed the chip slightly off so contact short.

It's not known if that was shipped to them damaged or they did it via user error, id be shocked if it was the exact same issue as last time as this was added to AMDs validation process for boards partners.
Click to expand...
Both cases so far, using the same board and have the same physical damage to the outside of the socket. At this stage, looks like PEBKAC.
 
LtMatt said:
@Robert896r1 @Poneros
Further tuning, based on the silicon and cooling equipment available to me. 64155 new PB score.


The next step to improve performance further would be to bring out the portable air con unit and press it against my 360 AIO, which I predict will add maybe a few more hundred points onto my score.

I saw Frame Chasers posted some results of the 9800X3D vs 14900KS in Riftbreaker. A game that is rather unfavourable to AMD CPUs for whatever reason.


I therefore decided to copy his settings running 'The Riftbraker' using my 24/7 tune profile, the results are below. Please note, this is not MAX Overclock.


My results were, 11% better 1% low FPS and 3% better average FPS than his 'MAX OC' results.
Click to expand...
Give it a try with max settings (incl RT), 4K, XeSS Perf. Absolutely brutal then. Curious to see if it's closer to 3x my fps for you.

BwaFFqC.png
 
Gentlemen, this represents the extent of my capabilities with these two benchmarks. Unless I revert to the 1.2.0.2 AGESA with its lower latency, these results are the best I can achieve for the foreseeable future. @Robert896r1 @Poneros




@Poneros - I'll give the Riftbreaker benchmark a try at some point and tag you when I'm done. :)
 
Last edited:
LtMatt said:
Gentlemen, this represents the extent of my capabilities with these two benchmarks. Unless I revert to the 1.2.0.2 AGESA with its slower latency, which is not in my plans, these results are the best I can achieve for the foreseeable future. @Robert896r1 @Poneros




@Poneros - I'll give the Riftbreaker benchmark a try at some point and tag you when I'm done. :)
Click to expand...
nice, you beat hero to 400 :D
 
You must log in or register to reply here.
Back
Top Bottom