Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror

Comment: Re:Many DDR3 modules? (Score 2) 136

by DigiShaman (#48666767) Attached to: Many DDR3 Modules Vulnerable To Bit Rot By a Simple Program

In my personal experience of "benchmark queens" in general; be it automotive performance or computing, are all about the synthetic numbers and zero basis on practicality (let alone value in cost). If a gamer is doesn't give a toss about a particular core subset of general computing (Video, CPU, RAM, and Storage), they're not benchmark queens. I've met plenty online who are. And when queens start debating online over numbers, the flamewars begin.

Comment: Re:good news for ECC memory makers (Score 2) 136

by DigiShaman (#48666601) Attached to: Many DDR3 Modules Vulnerable To Bit Rot By a Simple Program

Ouch! Seriously bad. Worse than the Pentium FPU bug (and that's bad). What good is a computer if you can't rely on the data being committed back to disk because of corruption mid-flight in RAM?! At least with the FPU bug, it was only FPU. But here we're talking about an industry wide issue where any operation cannot guaranty data doesn't become corrupted back to disk. By the time bit-rot sets in, you may have to dive into your grandfather-father-son backup archive. And that's assuming such a backup scheme is being used by those who are effected. Shit, that's assuming people are even backing up their data in the first place!!

Comment: Re:Many DDR3 modules? (Score 4, Interesting) 136

by DigiShaman (#48666555) Attached to: Many DDR3 Modules Vulnerable To Bit Rot By a Simple Program

True, and commodity chips not to exact spec will introduce disturbance errors. But apparently this is been a known problem with DRAM with various method of mitigation during the binning process. It's just that density and tolerances have become so tight that the issue is now exasperated. I wouldn't be surprised at all if those 19 models also had a few that failed if tested again and again.

Honest. General computing from low-end PCs, phones, and other devices are long overdue in employing ECC by default. So you lose capacity and tiny performance hit. BFD if that means your data doesn't become corrupted. The only people that would care are the PC gaming benchmark queens.

Comment: Re:Many DDR3 modules? (Score 4, Insightful) 136

by DigiShaman (#48666463) Attached to: Many DDR3 Modules Vulnerable To Bit Rot By a Simple Program

FTFP. "We induce errors in most DRAM modules (110 out of 129) from three major DRAM manufacturers."

Short version, leakage current from adjacent gates can nudge other to bit-flip. I don't think this is a manufacturing problem as it is a fundamental EE design oversight. So yeah, defective by design (unintentionally)!!

Comment: Re:ROM (Score 1) 163

by DigiShaman (#48662821) Attached to: Thunderbolt Rootkit Vector

This exploit can be used both ways as a "tool", right? If a malware infected Thunderbolt external drive can flash the EEPROM with a rootkit, is there any reason to believe that Apple couldn't create a utility Thunderbolt ROM drive (read-only to prevent client laptop cross-contamination) to stomp it back out?

"Your stupidity, Allen, is simply not up to par." -- Dave Mack (mack@inco.UUCP) "Yours is." -- Allen Gwinn (allen@sulaco.sigma.com), in alt.flame

Working...