Benchmark was performed on NEC/Renesas USB 3.0 controller uPD720200 firmware 3016. I've seen slight (~ +5%) performance variation for the same benchmarks on a different NEC USB3 equipped laptop.
Monday, April 23, 2012
Patriot SuperSonic Pulse
Some quick benchmarks for a Patriot SuperSonic Pulse 16GB USB3 (Product number: PSF16GSPUSB). As I suspected the reads are fast but writes are slow (sub USB2 speeds). This is not unexpected especially considering this flash disk was < 20$. It's also interesting to observe when running on USB2 the reads are faster than stock standard USB2 flash disks.
Benchmark was performed on NEC/Renesas USB 3.0 controller uPD720200 firmware 3016. I've seen slight (~ +5%) performance variation for the same benchmarks on a different NEC USB3 equipped laptop.
Benchmark was performed on NEC/Renesas USB 3.0 controller uPD720200 firmware 3016. I've seen slight (~ +5%) performance variation for the same benchmarks on a different NEC USB3 equipped laptop.
Monday, March 19, 2012
USB 3.0 Controller Firmware
Just to recap NEC/Renesas USB 3.0 controller uPD720200 firmware 3016 on my ASUS P6X58D Premium worked quite well with my Buffalo 1TB 2.5" HDD, delivering as expected results where finally the HDD was the bottleneck and not the USB connection itself. This is why I was surprised at the synthetic benchmark results for another Buffalo drive on the same PC, this time HD-LB2.0TU3-A3.
Initially contemplating this could be due to Western Digital internal drive within the enclosure (I've had a bad run with any WD disks for years now). Notice how slow and intermittent the reads are behaving compared to the writes. I suspect the writes were representative of disk's buffer & RAM caching more than the HDD speed itself.
Initially contemplating this could be due to Western Digital internal drive within the enclosure (I've had a bad run with any WD disks for years now). Notice how slow and intermittent the reads are behaving compared to the writes. I suspect the writes were representative of disk's buffer & RAM caching more than the HDD speed itself.
Results with USB 3.0 controller firmware 3016:
Baffled as I was, swapping USB 3.0 port connections around, checking for driver updates and rebooting nothing seemed to help. The internal HDD @ 5400RPM (= 90Hz pitch * 60 sec, aka WD's IntelliPower) should be able to give much better and more consistent sequential read results. I finally remembered how suss SSD controller firmware can get with particular disks and decided to upgrade the USB 3.0 controller firmware to version 3012 (with the same 64bit driver version 2.0.4.0).
Results with USB 3.0 controller firmware 3021:
I believe the results speak for themselves. Seems to me any high speed interface would benefit from a firmware upgrade these days. Plug-and-Play isn't what it used to be back when interfaces were running at much lower speeds...
Subscribe to:
Posts (Atom)
