Silicon Power Announces Ultra-Fast E20 SSD

Status
Not open for further replies.
[citation][nom]nebun[/nom]nothing new...SSD from crucial is faster[/citation]

But the Revodrive is faster.

Anyway, would have liked SP (Silicon Power) to have implemented an SF controller =P I won't mind though if they have price/gig lower than other competitors.
 

samdsox

Distinguished
Jan 3, 2010
382
0
18,810
Is it just me or is ssd starting to get like apple news on Tom's when the ipad came out. Tom's does so many articles about them . I think there needs to be a no ssd site like a no apple site. :)
 
[citation][nom]ehanger[/nom]Isn't there a PCIe SSD with read/write speeds close to 1GB/s?[/citation]

True, but. What price?

Revodrive 50GB $200
Crucial C300 128GB $275

Revodrive 960GB $3,359
SSD 1TB $2,530
 

dEAne

Distinguished
Dec 13, 2009
2,189
0
19,860
This is what we need:
"With E20’s built-in DDR2 DRAM Cache Memory, even during continuous intensive-write situations, computers shall lag no more," the company said. "This feature not only enhances SSD’s read and write performances, it also extends SSD’s lifespan."
 
G

Guest

Guest
Why stop at DDR2?

Why not use DDR5 instead for even better performance?
 
G

Guest

Guest
"The drive reportedly uses a proprietary technology that allows for read speeds up to 250 MB/s and write speeds up to 230 MB/s, surpassing similar SSD speeds on the market today."

Dont most SF1200 based drives do 270-285MB/S, and have been on the market for months. So it in fact is not surpassing similar SSD's and if it is turbocharged, then its a V6 going up against V8s.
 

maximus20895

Distinguished
Nov 28, 2009
54
0
18,640
Yes, most sandforce drives are 275-285MB/s write/read. Who cares anyway what the sequential is..the drives do 4k most of the time anyway.

I want a 6B/s intel controller.
 

g00fysmiley

Distinguished
Apr 30, 2010
2,175
0
19,860
SSD prices are actually continuing to go down. almost bought one this weekend at 120 gigs for 175 a good corsair

as mroe competition comes the prices will keep droppign and before you know it at least boot drives will be realistic

... maybe one day 1TB SSD's will be atainable at the $1 per gig and be down to $1k in stead of the current $4k
 
Sounds like a great drive but likely overpriced.
I think I might buy the Micro Center 64GB SSD, it uses the SandForce 1200 chips which allow for some of the fastest SSD speeds available today.
(It's only $99) Although I'd certainly like a higher capacity drive for the same price...
 

superblahman123

Distinguished
Mar 26, 2009
144
0
18,680
[citation][nom]Renegade_Warrior[/nom]Why stop at DDR2?Why not use DDR5 instead for even better performance?[/citation]

DDR5 is expensive and DDR2 is stable enough. Mind you, it's built in so the connection and speed should be great.
 

hrrmph

Distinguished
Oct 5, 2008
8
0
18,510
Yeah, the effect of the DDR is the big mystery here. Will it really work to improve things?

You would think it should help with responsiveness on short-term intensive tasks that saturate the ability of the NAND controller and NAND chips to respond. It's basically adding another layer of cache that should be faster than the (main) NAND chips.

If the DRAM controller has good logic and guesses correctly what the user needs to get from the drive most of the time, then this might be a winner.

On the other hand, if they are just out to say that they were the first to add DRAM cache, but do a sloppy implementation of it, then it will be useless.

This is what killed the first hybrid drives (sloppy implementation with too little NAND cache) and what makes the current Seagate Momentus XT hybrid drives a success, with good controller logic implementation with enough NAND cache to make a difference.

I hope Silicon Power gets the technology right and also offers the drives at a reasonable price. A premium for better performance is okay, but a steep premium would be a mistake in the ever more competitive and price sensitive SSD market.

-
 
G

Guest

Guest
1: What is the MLC count?
2: How many ECC bits? (will also tell me the MLC count)
3: What is the page and block size?
4: Is TRIM cmd supported from within VMs (the host OS will not support TRIM on VM image files)
5: Is write performance rate sustained? (large file write filling cache & not enough empty cells)
6: What is the average # of erase/write cycles before a cell fails?
7: How much $$
 
Status
Not open for further replies.