Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.sys.raspberry-pi > #9632

Re: A new (and better?) kid on the block?

From rickman <gnuarm@gmail.com>
Newsgroups comp.sys.raspberry-pi
Subject Re: A new (and better?) kid on the block?
Date 2015-09-13 11:45 -0400
Organization A noiseless patient Spider
Message-ID <mt45im$594$1@dont-email.me> (permalink)
References (1 earlier) <msh8kn$f8v$1@dont-email.me> <mshq3j$jbd$1@dont-email.me> <mshshk$spq$1@dont-email.me> <mt3n2r$den$1@dont-email.me> <ut5bva9hjd6voeipnqpbo44l0aihrps9f8@4ax.com>

Show all headers | View raw


On 9/13/2015 11:39 AM, Dennis Lee Bieber wrote:
> On Sun, 13 Sep 2015 12:37:46 +0100, druck <news@druck.org.uk> declaimed the
> following:
>
>
>> SD cards may have high headline transfer speeds for large sequential
>> transfers by they are far slower than a modest laptop drive for the
>> small random transfers which you get when using them as the primary disc
>> for an OS.
>>
> 	This would be the place to point out that Class 10 cards are rated
> explicitly for large single streaming transfers (video written to a freshly
> formatted card). Class 6/4/2 cards are still rated on smaller fragmented
> I/O (still image cameras with some images deleted, MP3 music files). While
> I would hope the top brand class 10 cards manage a class 6 (or at least 4)
> response on fragmented conditions, there is no promise of such.

I'm not clear on how that is a worst performance than a rotating hard 
drive.  When you have random access on a hard drive the seek time 
dominates the performance.  On writes this can be mitigated by a large 
buffer, but that won't work for all combinations.  For reads, random 
accesses are the worst for a rotating drive.  Performance drops like a 
rock.

-- 

Rick

Back to comp.sys.raspberry-pi | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

A new (and better?) kid on the block? "gareth" <no.spam@thank.you.invalid> - 2015-09-06 12:14 +0100
  Re: A new (and better?) kid on the block? "James Harris" <james.harris.1@gmail.com> - 2015-09-06 12:40 +0100
    Re: A new (and better?) kid on the block? "gareth" <no.spam@thank.you.invalid> - 2015-09-06 17:38 +0100
      Re: A new (and better?) kid on the block? Rob Morley <nospam@ntlworld.com> - 2015-09-06 18:13 +0100
        Re: A new (and better?) kid on the block? Andy Burns <usenet.feb2014@adslpipe.co.uk> - 2015-09-06 18:27 +0100
      Re: A new (and better?) kid on the block? "James Harris" <james.harris.1@gmail.com> - 2015-09-06 18:20 +0100
        Re: A new (and better?) kid on the block? Robert Riches <spamtrap42@jacob21819.net> - 2015-09-08 00:44 +0000
          Re: A new (and better?) kid on the block? Torfinn Ingolfsen <tingo@home.no> - 2015-09-08 18:18 +0200
            Re: A new (and better?) kid on the block? "James Harris" <james.harris.1@gmail.com> - 2015-09-08 18:23 +0100
        Re: A new (and better?) kid on the block? druck <news@druck.org.uk> - 2015-09-13 12:37 +0100
          Re: A new (and better?) kid on the block? Dennis Lee Bieber <wlfraed@ix.netcom.com> - 2015-09-13 11:39 -0400
            Re: A new (and better?) kid on the block? rickman <gnuarm@gmail.com> - 2015-09-13 11:45 -0400
              Re: A new (and better?) kid on the block? Dom <domafp@blueyonder.co.uk> - 2015-09-13 17:45 +0100

csiph-web