backtop


Print 25 comment(s) - last by winterspan.. on May 22 at 2:45 AM

Patriot has a opportunity to make a splash in the SSD market, if only you could buy their latest drives

Patriot Memory is a company best known for its DRAM modules, but as with a lot of their competitors, they've been trying to expand into the rapidly growing Solid State Drive (SSD) market.

The big fish in the enthusiast SSD market right now is OCZ Technology's Vertex series. Using a NAND flash controller from Indilinx, it offers superior random read and write capabilities, being surpassed only by Intel's more expensive offerings. They recently launched the Vertex EX series, a SLC flash version targeting the enterprise market.

However, having a premium product has led to high demand, resulting in a premium price. Other competitors have started launching SSDs using Indilinx controllers as well. Super Talent has the UltraDrive LE using SLC flash, and the UltraDrive ME using conventional MLC flash. G.Skill's Falcon series and Solidata's K5 and K6 series also use the coveted controller.

Patriot had originally announced its latest SSD on April 30 as the Fusion SSD series.  However, no drives were actually made available for sale. The company then renamed them as the Torqx SSD series, but as of publication none are currently available for purchase. Patriot assures us that the drives are now in mass production, and will be shipping out to distributors next week.
 
The Torqx series uses Indilinx's controller and comes with 64MB of onboard cache. Patriot is promising maximum sequential read speeds of up to 260MB/s and maximum sequential write speeds of up to 180MB/s for 256GB and 128GB models. The 64GB model is much slower, at 220MB/s max read and 135MB/s max write. Random read and write speeds were not available.

In order to win customers and gain market share, Patriot will ship free 3.5" brackets with all of their SSDs so customers can use their SSDs in desktops. Most SSDs come in the 2.5" format designed for laptops. This might be a good deal for consumers, but we will have to wait until the street prices of the Torqx series become available to be sure.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

No Random Read/Write numbers?
By erple2 on 5/20/2009 2:48:03 PM , Rating: 5
Is the reason why they never post random read/write numbers because they're abysmal? Are they any better than the first gen SSD controllers? Particularly on a "mostly full" drive? Has anyone actually tested each type to see what it's performance in real world scenarios are?

The sequential read/write is great for storing single large files, but the capacities of these SSD's is not conducive to that type of usage. I'm far more interested in how these things perform with tons of random read/write data.

Also, does the Samsung based drives provide respectable performance on a nearly full, or filled then erased (though not formatted) drive?




RE: No Random Read/Write numbers?
By Jansen (blog) on 5/20/2009 3:46:38 PM , Rating: 3
Since these use the Indilinx controller, I would expect them to perform similarly to Vertex SSDs. However, firmware could make a big difference.

I've been told that they will send a drive for Anand to test.


RE: No Random Read/Write numbers?
By someguy123 on 5/20/2009 9:55:45 PM , Rating: 2
Well, if they do meet the advertised sequential speeds the random writes should be taking a hit if they're using the same controller as the vertex.

The vertex is able to hit a few thousand I/O's, but at the cost of sequential writes. The high I/O firmware on the vertex drops the writes down to around 75mb/s.


RE: No Random Read/Write numbers?
By Chaser on 5/21/2009 9:09:54 AM , Rating: 2
Your sources for these random write speeds you're quoting on the Vertex drives?


By someguy123 on 5/21/2009 2:15:34 PM , Rating: 2
tony from the OCZ forums.


RE: No Random Read/Write numbers?
By barrychuck on 5/21/2009 9:27:32 AM , Rating: 3
Look on Newegg as I have posted a review of the Gskill Falcon drive that was full. That means I posted the worst case scenario after extensive usage, not just a fresh or "trimmed" drive. It still hit amazing random write numbers, almost beating out my hardware raid (highpoint 3510) and 4 x Gskill (Samsung) SLC SSDs in raid 0. The single Falcon can't beat the raid in throughput, but the IOs are impressive.

Also, Falcons are shipping with the same Firmware as the Vertex and post similiar numbers. This is because the firmware is from Indilinx not from OCZ. OCZ would have the customer's believe they tweek the firmware, but Indilinx has the greatest interest in improving the performance since they sell the controller to competitors. At that point, the flash itself is what matters.


RE: No Random Read/Write numbers?
By Chaser on 5/21/2009 9:58:42 AM , Rating: 2
I agree. I believe the random writes are far better than what is being stated here so far. thats why I asked for verifiable sources for these claims.

I have the Super Talent, Ultra drive ME 256GB, which also uses the same Indilinx controller. My random writes with Win 7 RC are quite good.


RE: No Random Read/Write numbers?
By Chaser on 5/21/2009 10:30:27 AM , Rating: 2
And I don't have the specific numbers as I am in a travel status right now and its on my home system but I ran every bench I could find in my shiney brand new SSD glee.


By winterspan on 5/22/2009 2:45:38 AM , Rating: 2
Everyone looking at the Vertex should now look at the OCZ "Summit" series which is still MLC but uses the new Samsung controller from the Corsair P256 with 128MB DRAM cache. it has better random write performance than the indilinx controller.

There is a pre-production review (Pcperspective.com I think) that showed it blowing away the Vertex drives.

The Intel x25-M may have the best seq. and random read speeds, but the Summit obliterates it on write speeds with >2.5X the performance.


RE: No Random Read/Write numbers?
By Quiescent on 5/20/2009 8:57:33 PM , Rating: 2
I'm not quite sure if there will be a difference between the flash SSD in my EeePC 4G Surf versus that sweet, sweet OCZ vertex 120GB SSD with 64mb of cache... But I do know that I've sat with below 60mb of space left on my 4GB SSD, and even right now, just 300mb of space reserved for Google Chrome, and I experience no loss of performance from the SSD. And to top that off, I haven't rebooted in 12 weeks, almost 13 weeks in less than 4-5 hours. ( Proof of uptime is here: http://uptimes-project.org/hosts/view/2332 )

That's why I'm so hooked on SSDs that I will definitely get one for my next desktop build. The stability of my 13 week uptime is 100x better than any long uptime I've seen on any OS on a harddrive. But yes, I'm getting off track here - My SSD has 1kb of cache if that, so says HDTune... Mmmm... These nice and decent SSDs with 64mb of cache... And get two and put them in raid 0... YUMMY!


By therealnickdanger on 5/21/2009 9:03:50 AM , Rating: 2
Three things:

1. Please don't use "yummy" again. Ever. *shudder*

2.
quote:
I'm not quite sure if there will be a difference between the flash SSD in my EeePC 4G Surf versus that sweet, sweet OCZ vertex 120GB SSD with 64mb of cache...

Was that sarcasm? IIRC, the EeePC SSDs have a max read of (at most) 25MB/sec. The Vertex is 10x that speed! However, it is doubtful that you could use the Vertex in the Surf anyway, making it a moot point.

3. I'm also not sure what SSDs have to do with uptime. While I can't prove it with a link (cool website, BTW), I've had plenty of HDD-based media servers over the years that have been up that long.


RE: No Random Read/Write numbers?
By Fat32 on 5/21/2009 9:38:48 AM , Rating: 2
quote:
Is the reason why they never post random read/write numbers because they're abysmal? Are they any better than the first gen SSD controllers? Particularly on a "mostly full" drive? Has anyone actually tested each type to see what it's performance in real world scenarios are?


Great idea. I tested my almost full Vertex 30Gb vs WD1001FALS using 4K random read/write of Crystal Disk Mark. Vertex was 7 times faster in reads and 4 times faster in writes. So that justified SSD performance/price over HDD for me.

When testing SSD vs SSD that's of course different story. But right now I don't want to spend a lot on SLC model.


"Google fired a shot heard 'round the world, and now a second American company has answered the call to defend the rights of the Chinese people." -- Rep. Christopher H. Smith (R-N.J.)














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki