Thunderbolt-Equipped Gigabyte Z77X-UP5TH Meets Velociraptor Duo Review

Jos

Posts: 3,073   +97
Staff
Read the full article at:
[newwindow=https://www.techspot.com/review/584-thunderbolt-external-storage/]https://www.techspot.com/review/584-thunderbolt-external-storage/[/newwindow]

Please leave your feedback here.
 
Another "thunder bolster". Such a useless technology. When USB3 and SATA-III aren't enough, Thunderbolt can't help you either, one goes for PCI-Express cards.

Plus, 10GBit Ethernet cards are getting cheaper, becoming a viable option for a fast network.

When I'm getting a new rig in a few months, I will opt for 10GBit Ethernet cards.
 
This is a huge disappointment. I was hoping to see some superior, if not equivalent performance as sata6. On top of that, the price...holy crap!!
An OCZ RevoDrive is currently $499 on NewEgg. It has a 480gb capacity, but the read speed is 1000mb/sec. The RevoDrive is only optimized for PCI-e 2.0 at this minute. If a RevoDrive could take full advantage of PCI-e 3.0, I would LOVE to see those read/write speeds. Maybe people should invest in optimizing the current technology instead of inventing new things that are ungodly expensive and inferior.
 
An OCZ RevoDrive is currently $499 on NewEgg. It has a 480gb capacity, but the read speed is 1000mb/sec. The RevoDrive is only optimized for PCI-e 2.0 at this minute. If a RevoDrive could take full advantage of PCI-e 3.0, I would LOVE to see those read/write speeds.

I don't think SSD's can even begin to saturate the bandwidth PCI-e 2.0 offers.
 
An OCZ RevoDrive is currently $499 on NewEgg. It has a 480gb capacity, but the read speed is 1000mb/sec. The RevoDrive is only optimized for PCI-e 2.0 at this minute. If a RevoDrive could take full advantage of PCI-e 3.0, I would LOVE to see those read/write speeds.

I don't think SSD's can even begin to saturate the bandwidth PCI-e 2.0 offers.
I'm sure that you can squeeze more bandwidth from a PCI-e 3.0 since you have more parallel information streams. Regardless, the industry needs to focus on optimizing what we have, not an investment into a new/risky architecture. Thunderbolt is high risk/high reward. Either it's going to completely flop or they will be rich. There will be no in between.
 
Regardless, the industry needs to focus on optimizing what we have, not an investment into a new/risky architecture.
Wasn't Thunderbolt(aka: Light Peak) intended as a replacement to all the antiquated interfaces that has been optimized over the top for the last decade. I personally think its time for a new standard of interfacing with PC hardware.

Maybe I'm confused again but wasn't fiber optics and 100G data transmission the initial goal for Light Peak?
 
I don't think SSD's can even begin to saturate the bandwidth PCI-e 2.0 offers.
PCIe 2.0 x4 slot has bandwidth of 2GB/s in each direction. Revodrive 3 X2 960GB has 1500MB/s read, 1300MB/s write peaks. If they were to replace the Sandforce controllers with the Everest 2 controllers, they have theoretically DOUBLE the IOPS. So that would comfortably saturate a PCIe 2.0 4x slot.

Edit: Why would I point out a PCIe x4 slot? Well unless you go to an gaming enthusiast board, Intel motherboards run one or two x16 slots then other slots like a x4 slot or two. If you actually use the second x16 slot, it usually drops both x16 slots to x8 so your graphics card gets bandwidth starved (this is noticable on any decent resolution like 1080p or in a hardcore graphically intensive game). So you really want to be using the x4 slot instead.

It is pretty simple to upgrade an SSD controller to up the bandwidth. Just double the bus widths and you double the bandwidth. Doesn't have anywhere near the limitations of a mechanical spinning disk.
 
Back