|
Silicon Power 2.5inch Solid State Disk 64GB |
|
|
|
Written by Ivan Vujic
|
|
Monday, 22 December 2008 |
|
Page 1 of 6  First thing that crosses everyone’s mind upon mention of SSD is high price or, lower price that results in lower performances (much lower even compared with Hard Disk Drives). It doesn’t have to be always like that and many manufacturers are trying to change that prejudice when it comes to SSDs. One of those companies is Silicon Power from which we got one MLC SSD 64GB with low price (for an SSD) and promising specifications.
From specifications we saw that read speeds are not that impressive as writing speeds are - 95MB/s. Of course there are faster SSDs, like Intel SSD with price tag that is almost 4 times higher than price tag for Silicon Power model, but it had slower write speed (70MB/s) but faster read speed. Another thing that caught our eye was MLC mark on package which means that this SSD uses Multi Level Cells and because of that it is slower compared to its SLC (Single Level Cell) cousin. Because of this life expectancy (durability) is shorter and power consumption higher but on the other side price is very acceptable. Average user shouldn’t be worried with shorter durability or higher power consumption because these devices will become cheaper over time so they will be replaced before they reach end of durability period. We found in package, beside SSD itself, short manual, screws for mounting and no cables and no adapters for 3.5” slot (standard HDD size for desktop cases) which suggests that it is intended to be used for notebook/netbook installation.
|
|
|
Testing procedures
| |
This SSD was tested on ICH 8, ICH9R and ICH10R South Bridge which have relatively new integrated controllers with great performances for average user. If you have relatively older ICH7 controller that was used on 945 chipset (lot of notebooks and netbooks are based on this one) you can expect slightly lower performance levels but real differences would be even smaller. |
To establish how this disk performs under real work conditions, besides synthetic benchmarks, we have done some test that can be used as real work reference: copying large amount of data and OS boot-up time. Tests were conducted on two desktops and one notebook. Desktop configurations were needed because we used different controllers to determine if user of older motherboard would get drastically worse performances. Configuration with Quad CPU was freshly installed on SSD and in two cases we used installations that were one year old (transferred with ghost). We compared SSD with RAID system, and one older hard disk because those are still present in many configurations. Every test was conducted two times and average value is used to minimize possible error.
|
