Sony 128GB G XQD card - not seeing the r/w speeds I would expect

[LJR]

I connected my XQD reader (https://biSPAM/2Q3nAPb) to my i7 iMac's USB 3 port (at the rear), inserted the Sony G XQD card and then ran Xbench and Blackmagic Disk Speed Test. I see a read speed of around 300 MB/s and a write speed (the important one!) of only 150 MB/s. While I don't expect to get the R:440MB/s and W:400MB/s reported by the card, I would expect it to be much closer than it is.

Any thoughts as to why it's so slow? What are other people seeing? Particularly interested in those of you connecting to USB 3 port on the iMac.

 

[swanseamale47]

I'm no expert but I wonder if the speed is limited by the USB rather than the card?
The other thought is it's a fake card, theres loads about, even in reputable stores.
Just my thoughts.

 

[LJR]

I looked up the USB 3 bus speed and it's 650MB/s, so I think we're good there. Fake cards! Yikes. I know that's a problem with SDXCs but wasn't aware that the Sony XQD's were being faked. I wonder how one can tell?

 

[Deleted sulking member 63079]

That adapter only has a theoretical max of 500. No mention of expected sustained rate.

And you have to factor in the read/write of the source/destination device too

 

[LJR]

That adapter only has a theoretical max of 500. No mention of expected sustained rate.

And you have to factor in the read/write of the source/destination device too

Hmm...yes. 'Max rate' would imply it's peak performance, as opposed to sustained performance when transferring a large data payload (5GB in my test)

I would assume the R/W of the destination, when compliant with USB 3.1, would be 650MB/s (as per the spec)?

Anyway, it seems like there are too many variables to do any kind of real test.

 

[Deleted sulking member 63079]

Hmm...yes. 'Max rate' would imply it's peak performance, as opposed to sustained performance when transferring a large data payload (5GB in my test)

I would assume the R/W of the destination, when compliant with USB 3.1, would be 650MB/s (as per the spec)?

Anyway, it seems like there are too many variables to do any kind of real test.

Depends if it's a physical HDD, SSD or what at the other end

 

[HoppyUK]

Card speeds quoted by manufacturers are theoretical maximums under optimum laboratory test conditions. Real world speeds never get anywhere near them. If you have another card of known quality, relative speeds should roughly compare according to spec (assuming there's no bottleneck).

Of all the cards likely to be faked, I'd guess that XQD would be one of the least likely. The best way to avoid fakes is to buy from a dealer supplied by the official importer.