Remember to always read at least twice, then compare the md5s before flashing
Many SOIC8 clips can give an intermittent/bad connection. You will only realize it when you try to restore and ... oops!
Many SOIC8 clips can give an intermittent/bad connection. You will only realize it when you try to restore and ... oops!
Setting a lower SPI speed didn't help. It felt like there must be a flaw in the Linux SPI implementation (/BeagleBoneBlack driver) that even though I lowered the clock speed, the delay between data and clock was still tight. Either that or the signal times were so fast, they were coupling enough to override an active driver?? Rather than setting up the scope to investigate, I just tidied up the leads.
Also, specific to Intel powered motherboards newer than a certain generation, the image should show as an "Intel PCH File Descriptor" or something if you run `file` against it. Don't use this as an integrity check though, `file` is just looking at an Intel-specific header at the beginning of the ROM file.
When you are flashing a bios as the other poster mentioned, you want to be sure you have the right file and you will be able to restore it.
About SOIC8 clips, if you can't get a reliable read, consider soldering with stripping wire straight to the IO ports.
Stripping wire is thin enough that you can't break a leg of the chip.
While trying to seat/reseat on the package, the lower pins in the Pomona clips would tend to get pushed in, bent, and pop out of their channels.
The cheapest clips, with the pins that go straight through, however, didn't seem to do this. To make space to attach individual header hookup wires to the pins at the top of the cheap clip, I bent the pin tops left and right, in an alternating pattern.