I’m gonna guess mining.

Tronics Fix might be your guy. He does electronics repairs, which sounds like it’s at least related to what you’re looking for.

Other than that sort of repair work, what you’re describing is just the basics of hardware troubleshooting. Identify potential points of failure and swap them out for known-good units until you figure out what caused the problem. From there most people (and even most shops) will just replace the faulty component either from their stock or with an RMA.

Event viewer and logs like that often times take more time than they’re worth when you can just swap parts one by one in some logical order until you hit the solution. I used to do repair/troubleshooting as a side gig when I worked at Gigabyte doing board design, and my steps whenever possible were to just swap parts for known-goods. I only got out the logging tools if something was super weird and rare or I couldn’t source a component to swap quickly, which was also rare as I collected a lot of different hardware over time.

Beyond that I might try to repair a broken board if it’s something simple like a broken header or port or something, but if you ever saw me doing a whole VRM swap or BGA soldering work, it was probably to revive something special, or too expensive to just replace.

It’s not storage, but memory. Storage refers to your hard drive or SSD. Memory is where processors pit data to work on.

In your case the iGPU is reserving a portion of your system RAM, likely between 128MB and 1GB. Windows will report this as “hardware reserved.”

You can probably change or at least view how much it is set to reserve in the BIOS. For example, my laptop with 32GB of RAM and a Ryzen 9 6980HS (radeon 680M graphics) reports 2.1GB as hardware reserved and the BIOS is set to allow 2048MiB (2GiB or about 2.1GB) for the iGPU to use.

Currently, the node advantage difference means Apple will be ahead on power efficiency, but Intel chips can and do outpace Apple’s right now.

The 1035g1 is 2 generations newer for the same product teir. It’s a 10th-gen i5 vs an 8th-gen one in your comparison.

You can view a comparison of specs for both here.

MTL is not currently planned for a desktop release.

Good thing this is Xe-LPG then. LPG is Alchemist based and goes into Meteor Lake. LPG+ goes into Arrow Lake.

Fixed!

The issue is that it breaks the commonly accepted trend of bigger number better. You could have an 8520U get clapped by a 7540U for example, because even though the latter is newer, the former is Zen 4 to its Zen 2.

The 9850H is going to crush the 6820HQ in anything CPU intensive. Your GPU won’t matter much for programming, but the former laptop being several years newer should be faster there as well. (Both integrated only or both dedicated).

If the screen size is a deal breaker, then do for that, but if you want the faster machine, the 9850H is significantly ahead. This would have tangible benefits in things like compile times if that is something you deal with frequently.

It should be a drop-in cpu for any sapphire rapids compatible motherboard, but if the bios can boot it is another story.

It may not perform like a production CPU, as clocks on ESes are often lower than production samples as defects get ironed out.

Either way, be careful with those! Somebody was not supposed to do that if they’re working CPUs. Even if they’re dead, they should probably stay internal.

Grade 5 can make more than that. You’re getting lowballed. Sadly, I expect nothing less from this industry. Somebody else dropped a link already for the typical ranges.

Reduce your undervolt slightly. -120mv is possibly stable enough given it lasted a few minutes.

How a GPU is broken down usually determines how other things are shared between those ALUs. I’ll use ARC Alchemist for this because I have the spec sheet for it.

The A770 is broken down into 32 Xe Cores. This means it has 4096 shading units, 256 TMUs, 128 ROPs, 512 Execution Units, 512 Tensor Cores, 32 RT Cores, and 16MB of L2 cache.

You can also think of this as each Xe Core being made of 128 shading units, 8 TMUs, 4 ROPs, 16 Execution Units, 16 Tensor Cores, 1 RT Cores, and 512KB of L2 cache.

That Xe Core is the smallest unit you could break an Alchemist GPU into and still have every part of the larger whole. You can’t literally just do that, cut the GPU in half, but if I had to draw a diagram of one that is what would be in each Xe Core block.

I’m not going to get into the technical side of how GPU design works, partly because that’s an entire doctorate thesis to write out, and also because I work on the CPU side and those guys are wizards to me.