Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Burying some way more interesting unnoticed metric in the logs.

In practice, I would trade a ~75% faster drive for <10% of storage in a thumb drive use case with zero hesitation:

Old: Probe time: 1'49" Read: 463.8ms / 4814 = 96us Write: 1'48" / 4192321 = 25us

New: Probe time: 1'02" Read: 338.2ms / 4814 = 70us Write: 1'02" / 4192321 = 14us



So, guess what, advertising that it's a bit smaller and much faster is probably the best way to get people like you to buy.


Was thinking the same, but with it not being called out in the article at all, I was wondering whether to trust these numbers. The tool says "probe", not "benchmark". Not sure what to make of it


I might too, but only if I knew that's what I was doing.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: