Ext. Enclosure Burn Tests on Nine Drives.

With all of the problems with external enclosures, I decided to do some testing and post the results. I took a Verbatim MCC 03RG20 16X –R discs and used it for all simulated burns. I also used a burned –R disc to measure the burst rate. Only one Liteon burn had any background activity, and it was minimal.

For those who want a summary, I list it below. For those who want to see the burn graphs, they are also posted.

All burns were on Plumax 525 series enclosures. The first were utilizing the Cypress AT2+ chip connected by USB2. The second was the Prolific 3507 chip with the 112805B firmware, connected through firewire. The last burns were with the same Prolific case connected through USB2. Links to the exact cases and firmware are at the bottom of the post.

http://www.dealsonic.com/plpmblcousb22.html

http://www.dealsonic.com/pl52usb20enf.html


http://www.prolific.com.tw/eng/downloads.asp?ID=44

First Burn: Cypress USB. Second Burn: Prolific Firewire. Last Burn: Prolific USB

First Burn: Cypress USB. Second Burn: Prolific Firewire. Last Burn: Prolific USB.

Second Set.

First Burn: Cypress USB. Second Burn: Prolific Firewire. Last Burn: Prolific USB.

Third Set

First Burn: Prolific Firewire. Second Burn: Prolific USB.

Fourth Set

First Burn: Cypress USB. Second Burn: Prolific Firewire. Last Burn: Prolific USB.

Fifth Set

First Burn: Cypress USB. Second Burn: Prolific Firewire. Last Burn: Prolific USB.

Sixth Set

First Burn: Cypress USB. Second Burn: Prolific Firewire. Last Burn: Prolific USB.

Seventh Set

First Burn: Cypress USB. Second Burn: Prolific Firewire. Last Burn: Prolific USB.

Eighth Set

First Burn: Cypress USB. Second Burn: Prolific Firewire. Last Burn: Prolific USB.

Ninth Set

Thanks for all the test burns. So it seems like unless the drive is UDMA-4, it can't achieve 16x burn through the external enclosure. The oddity is the LG 4166, using the Panasonic chipset, which seems to be a bit more tolerant.

Thanks chas0039 for all work you put down on this test.
Hope somebody can make this thread sticky.

... I was living under impression LG 4166 also runs in UDMA-4 mode. Apparently not.

Yes, this is why I just can not understand why all writers are not udma4!

BTW plextor uses udma4 also so thier speeds would be comparable to the liteons.

Also from what I have tested the oxford 911 is comparable to the firewire prolific chipset in terms of speed. The 911+ is about 1 mbs faster than the prolific in firewire mode.

Chas did you have issues with your pioneer 111 and the new cypress chipset?

Also Chas what was the specs on the computer you tested this on?

The Pioneer 111 would just not recognize the blank disc when hooked up to the Cypress but it would do the burst test with the burned disc. I could try it again if anyone needed absolute confirmation.

The computer is a Pentium 4, 2.4 Meghz, PC 3200 Dual Channel 1Gig RAM, with 3 SATA hard drives. Relatively slow, but no constrictions that should have limited the drive speed.

Hi Chas,

great work
The performance of a drive in an external enclosure is also dependant on the USB- or Firewire-controller of the host computer. So it would be fine, if you'd mention these components used in your test.

Michael

Chas, thanks for all the scans. I think there's a prolific enclosure in my future now...

Hi,why? According to the table, the Cypress seems to be more consistent, unless you want to connect a Pioneer.

Michael

USB is onboard an Intel 865 PE MB. Firewire is an add-on card with no information.

Hi,
Then have a look into device manager. That shows at least the manufacturer of the chipset. Firewire controllers are normally not integrated into the chipset, but additional components connected via PCI, even if they are soldered onto the mainboard.

Michael

I did and there is no information other than OHCI Compliant. Maybe this is what you want; I assumed it just meant "compatible".

I've got a 165P6S that I'll be connecting via firewire...

Yes,thanks for all this work !! It will help me when choosing as enclosure for sure !

g

Thanks for the tests Chas!

chas0039 be assured, your system is fast enough for a test like this one.
You'll be able to read (manufacturer/model) information on PCI cards chip.
The same goes for enclosures FW chips.

Thanks again.