I’m willing to listen to what you say (no problem to be expected)… It just goes against many years of ‘standard procedure’, if you know what I mean…
I Must Say… I was expecting ~350R/W due to my controller being an embedded SATA II controller (Supermicro X7DA8+) and I’m getting far more: 510R/495W using the AJA Speedtest program set for 1920x1080 at 8 bit resolution.
If what you say is true… I can see a couple of spinning Raid 0 arrays being exchanged for SSD’s in the future…
But ignoring 20+ years of ‘procedure’ is going to be a hard thing to ignore… I referred to it as: “Gag Reflex” in another thread…
I beat the living heck out of the edit I bought the new raid for, with “test revisions” where I threw layer after layer at it to try and break it… and couldn’t break it… (Dropped Frames indicating that drive couldn’t ‘keep up’ ) Which is very promising!
Raid 0 for video editing is primarly write once or twice, read many… As when revisions and effects are created, they are stored in different locations, on same or other drives… So Video Editing is not anywhere as write intensive as say a database (which does exist in video editing: usually on a different drive than media so as to distrubute the load on the available disks )
If you have any other information, or ‘white papers’ on the subject… or speed tweaks that I could implement…
I would love to read more into it…
Jan
