IRC log of #cubox of Sat 20 Jul 2013. All times are in CEST < Back to index

00:43 dv_ MMlosh: any progress on the sata speed ?
00:44 MMlosh it stayed at 68MBps = 544mbps during the transfer, as I said earlier
00:51 TrevorH1 esata attached here: Timing buffered disk reads: 302 MB in 3.00 seconds = 100.64 MB/sec
00:52 dv_ TrevorH1: what did you use for measuring?
00:52 dv_ copied with dd ?
00:52 MMlosh I am currently trying to copy a file using a faulty SD card reader using ddrescue
00:52 MMlosh it's fun
00:52 TrevorH1 that's hdparm
00:53 dv_ i dont know which sata version the cubox uses
00:53 dv_ but even sata 1 can theoretically do 150 MB/sec
00:53 dv_ it would be interesting to see what the cubox can do with an SSD
00:54 TrevorH1 ata1: SATA link up 3.0 Gbps
00:55 TrevorH1 that's what mine is
00:56 MMlosh ouch.. but I think I got my 4GB file out of the card..
00:56 MMlosh (I had to press the card rather strongly inside the reader for the whole time)
00:56 MMlosh I remember mine also saying 3Gbps
00:58 MMlosh my hdparm -t says 174 MB in 3.00 seconds = 57.99 MB/sec
00:59 MMlosh it could simply be the drive. It's noisy and the capacity is only around 250GB.
00:59 TrevorH1 weirdly my ssd is quicker at writing than reading
01:00 dv_ I think I have heard about such a phenomenon
01:00 TrevorH1 ah, wait, bloody debian has /tmp on tmpfs ;)
01:00 MMlosh I have a SD card by Kingston here.. it does >150MBps writing, but only 30MBps reading.....
01:01 MMlosh don't forget that for small data quantities & using dd for measurements caches come into play
01:01 TrevorH1 it's still quicker but not that much
01:02 TrevorH1 dd if=/dev/zero of=/home/trevor/file bs=1M count=1024 oflag=direct --> 1073741824 bytes (1.1 GB) copied, 8.80006 s, 122 MB/s
01:05 TrevorH1 using conv=fdatasync instead of oflag=direct gives a more expected result of ~75M/s
02:40 dv_ \o/ vc-1 playback works now
02:48 Coburn Makes me wonder if I should go back to CuBox dev
02:48 Coburn I've been playing with R-Pis too much :P