Announcement

Collapse
No announcement yet.

Crucial BX100 SSD Tests On Ubuntu Linux

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Crucial BX100 SSD Tests On Ubuntu Linux

    Phoronix: Crucial BX100 SSD Tests On Ubuntu Linux

    One of the latest solid-state drives at Phoronix that's been up for testing is the Crucial BX100, a mid-range SSD with the 120GB version retailing for around $60 USD.

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    While the Vertex 3 is a few years old now, it was a higher-end solid-state drive and can still beat out the cheaper SSD competition of today.
    The Sandforce SF-2281 chipset the Vertex3 is based on performs on-line compression of data written to it - and because most benchmarks simply write zeros, performance seems to be a lot higher than it actually would be in real-world workload. "Real" hardware review sites are actually aware of this issue and at least mention it somewhere, instead of blindly trusting some benchmak results

    Comment


    • #3
      Originally posted by Linuxhippy View Post

      The Sandforce SF-2281 chipset the Vertex3 is based on performs on-line compression of data written to it - and because most benchmarks simply write zeros, performance seems to be a lot higher than it actually would be in real-world workload. "Real" hardware review sites are actually aware of this issue and at least mention it somewhere, instead of blindly trusting some benchmak results
      Yeah, I figured the Vertex was 'cheating' somehow ... the discrepancies were just too glaring.

      Comment


      • #4
        This drive is on the queued trim blacklist(along with all m5* and samsung 8*), but apparently the mx200 is not.
        I tend to base my purchase decisions on things like this, or more importantly if I can update the fw without buying windows (crucial was the first to allow this).
        Last edited by Soul_keeper; 25 September 2015, 11:21 PM.

        Comment


        • #5
          Originally posted by CrystalGamma View Post

          Yeah, I figured the Vertex was 'cheating' somehow ... the discrepancies were just too glaring.
          Are Vertex made by VW ?

          Comment


          • #6
            Originally posted by Soul_keeper View Post
            [...]`I tend to base my purchase decisions on things like this, or more importantly if I can update the fw without buying windows (crucial was the first to allow this).
            Do you perhaps know which other ssd manufacturer will let me upgrade fw from Linux? Samsung? Cheers

            Comment


            • #7
              I'm kind of out-of-the-loop on SSDs. If I do a lot of (Compressed) Video Recording and editing, would I be OK with one of these? I was looking to buy maybe a 256GB SSD.

              Comment


              • #8
                I've been eyeing this one for some time now, but I realized that in my country, the difference between the BX100 and MX200 is 20 euro, so it makes more sense to go with the MX instead. Might grab one on next paycheck. Just have to convince myself I really need it first (I don't )

                Comment


                • #9
                  Originally posted by Soul_keeper View Post
                  This drive is on the queued trim blacklist(along with all m5* and samsung 8*), but apparently the mx200 is not.
                  Where do you see this? Looks like the drive simply does not support queued trim, which should be fine. It's not on the blacklist, in any case:

                  Linux kernel source tree. Contribute to torvalds/linux development by creating an account on GitHub.

                  Comment


                  • #10
                    Originally posted by Linuxhippy View Post

                    The Sandforce SF-2281 chipset the Vertex3 is based on performs on-line compression of data written to it - and because most benchmarks simply write zeros, performance seems to be a lot higher than it actually would be in real-world workload. "Real" hardware review sites are actually aware of this issue and at least mention it somewhere, instead of blindly trusting some benchmak results
                    That is simply not true. Benchmarks either use real-world (like) data or they have separate compressible and incompressible data tests.

                    Comment

                    Working...
                    X