GzeroD

My Guesses would be a driver issue causing a resource allocation issue if its not that the issue may be might be your motherboards stability on its PCIE lanes or interference from the new GPU. I would try the following. Update chipset and GPU driver, Clean install if you can take the time for the extra restarts. If that does not work I would look at Try busting the GPU down to 8X speeds and Gen 3 in bios to check to see if it's more stable,, as that may help if it is one of the stranger resouce allocation issues.

For most games the Ryzen 9 is looking to be faster by quite a bit in some games but on average they are quite close Ignore the Simulated result for the 7800X3d as I dont know if Hardware unboxed adjusted for the large clock difference between the 7800x3d and the 7950x3d that they disabled on od the ccd's for in the test.

That is going to depend on exactly what you're trying to do.

As I said in my post the pins are very fragile. Just bending them back and forth 5 to 10 times should have them snap at or near the solder joint. I would not use a razor blade as you could easily nick traces used for other parts of the motherboard and brick it. Scissors could trim the. Low enough so that you can cover the bits left with electrical tape and be reasonably safe from any problems.

Would seem it's not shorted right now. that said something touching those pins while the pc is plugged in or the caps on the board are still charged it could fry the PC or one of the components. If that PCIe slot is not important to you I would just unplug the PSU, press the power button so the fans spin for a second and hopefully clear all the caps, then after unplugging the CPU power connector and the motherboard 24 pin connector get rid of the pins. it sucks to lose the PCIe slot but it would suck a lot more for a falling cable or just a pin drooping over time to short and nuke your PC and some/all the parts.

Assumeing it cant be RMAed. If you have confidence I would bend the pins back in position and gently but firmly push the PCI express slot back onto the pins. Then I would check to make sure the contacts could be seen in the slot, if there are any spots where you cant see a pin its probably stacked on top of another and turning on the system like that could fry it. If you don't have confidence in that or they there are broken pins already I would just cut them off so they can't short when you power the PC back on and just lose the PCIe slot. Side cutting or flush cutting pliers would make quick work of them.If you don't have those just bending the pins back and forth a few times should break them as they are very fragile.

I would attempt a DDU on the 3090 driver and do a fresh reinstall of the driver after a reboot. I'm guessing the chipset driver updated something for your cpu's IGP/driver and it clashed with your already installed GPU driver.

I would attempt the Bios Flashback again. Generally a Flashback wont require step by step updates that manual flashing from inside the bios might But I have seen several cases where they where very picky on the Drive they would allow to flashback from. I generally use an old PNY 1 GB usb 2.0 drive as it fails the least. If you're unable to get the board to work on a reflash attempt with the newest bios, I would use an older version of the bios that supports the 13th Gen CPUs Generally the first few available are all but certain to work with flashback to get a new CPU working. That said you may want to only use 1 stick of memory (in different slots if the forst fails) and the CPU with nothing else plugged in too see if you can get into the bios after the flashback. This is to check too see if the board has other defects. There could be a some bad traces causing the boot issue with certain slots or items installed/populated. and the board was on woot so it will most likley be a refurb which in my mind makes that more likley.

Here is a screen shot of the HDD service section of the R51 Thinkpad Manual. The complete manual can be found here https://thinkpads.com/support/hmm/hmm_pdf/39t6190.pdf .

Answering this well really comes down to your definition of "Large data sets" and "lots of ram". Given what you've mentioned I would Imagine that 128GB is baseline requirement as just holding a few compiled Regular expressions in memory can be several GB of RAM on arrays that are only a couple hundred thousand entries. With that in mind depending on the size of your expected datasets> You may easily end up on a prosumer platform because of the limitations of how much ram can be added. Also are you sure no GPU or accelerators can be used to help crunch the data you're going through? Sometimes it shocks me to see how fast a CUDA accelerated task will destroy my R9 5950 working through the same problem with a non-cuda enabled version even though I'm just crunching through it with an old M40 I was able to get on eBay for less than $35 early last year. If you have any tasks that may be accelerated with an accelerator grabbing one from a few generations ago will often make sense even if you only use them once in a great while (generally opting for the largest VRAM options to enable you to do as many possible datasets in your price range.) Before we can get into the nuts and bolts of a solid recommendation, if we don't know if your workload requires a workstation platform we can only give half hearted recommendations.

Moved this over to the troubleshooting section.

Let the dead lie. Thread Necro Locked.

This. Could we get the name of the case you are using or a picture of the inside of it. Or better yet a link to the prebuild from the vendor.

You could try installing speedfan and forcing the fan to run faster, assuming it is supported on your laptop.

Minor Cleanup In Progress. Cleanup Complete Keep it Civil