SSD's get me excited for some reason. They're getting faster, evolving, getting cheaper, are mysterious, yet Anandtech peels away the mystery.
I can't get enough.
I can't wait to get my Win7 pro, and SSD in October. I'll decide in October which SSD, but Intel just made it sweeter.
One interesting but ultimately irrelevant derived statistic from the G rating of this drive...
It can be accelerated to 7.35km/s over that 0.5 seconds.... discounting drag that's the delta-v required for low earth orbit....
Unfortunately that would require a canon/rail over 3km long ;']
I would like to kill (read: eliminate from existence) the tech writer, probably from HardOCP (read: guy who likes to review power supplies and use the non-word "immersive" way too much), who started using this "read" (read: read) terminology to explain what he meant.
I haven't ever really tried to purchase brand-spanking new hardware before, so one question - how long does it usually take for newegg to start selling new hardware?
It is impossible for car makers to do this!
Stop comparing APPLES to ORANGES.
People have paid through the nose for Intel drives and now Intel have left them bloody.
It's a simple fix that Intel are CHOOSING to do, and yes that is why people are feeling screwed.
I bet Intel would have to steal your wife and kill your dog before you said a bad word about them!!!
I forgot to mention, doesn't 150mw seems very low? Other site has been reporting other SSD using up to 4W when active. Which means it aren't saving much energy compare to 2.5" HDD.
May be anand could do a power table comparison when you ever decide to do an SSD roundup.
I was hoping you could point to a link that states this from Intel. My x25-m is fairly new (less than a week) and I'd like to make sure that if I was to return it that it was because TRIM isn't supported on this drive.
The read/write stuff I can live without, but extending the life of the drive is what I'm down with.
Since SSD lack rotating parts and are pretty much unsensitive to dust I can't really see the reason to continue with an enclosure and the accompanying cable harness that is the case with a 2.5"/SATA-drive. And how long will it take before even SATA 6Gbit is saturated performance wise, 2-3 years?
I would really like to see a PCI-express version of this and other new cheap performance oriented SSD.
Would be interesting to here your word on this Anand.
Best regards
The one thing I was wondering about SSD's was how good they would be for the indexes on a newgroup reader> Specifically using Agent version 5 and putting it on an SSD by it self and letting the reader grab over 40G of data. Then when it does the purge and compact on the databases I wondered how it would stack up with a typical hard drive. Does anyone have any ideas if this would be a good use or not??
Thanks!
Since Intel doesn't appear to be introducing new SLC Extreme drives in the near future, any word of a firmware update for the 50nm 25-E drives to include TRIM support when Win7 is released in October?
My (limited) understanding is that it's not quite as necessary for the TRIM command for an SLC drive. However, it seems to be that it couldn't hurt, right?
With that being the case, any idea if Intel will be working that into the firmware for the x25-E, since they don't plan to introduce a 34nm model?
"The new controller is Halogen-free (the old one wasn’t) so Apple could theoretically use the new drives in their systems without being un-green."
Yeah, because it is very important that Apple "feels" green, after their Mac pro burns off ~.3KW's worth of power. Theirs are not the only systems that use "too much power" though.
One of these days, OEMs will get a clue, and perhaps then push the industry into using truly "green" components. It is great that we now have RoHS ( although solder is a mixed bag if it does not bond right, or well ), and low power usage components such as the SSD. Maybe some day CPU/GPU manufactures will get a clue ? Sure, *someone* needs that 1kW space heater of a PC. Now we need something that can do most tasks without forcing the user to buy a laptop/netbook, and using enough energy to power 2-3 mid sized refrigerators.
Any word on whether the drives are ONFI 1.0 or 2.1? I'm assuming, based on the speeds, that it's (unfortunately, and strangely) the former. I'm disappointed that Intel didn't use the new Micron 34nm ONFI 2.1 NAND.
I'm not referring to the RAM cache - I'm referring to the NAND used for storage. Intel partnered with Micron to form IM flash technologies, and Intel uses the NAND from IM in their SSDs.
No, the most important thing for drives of this size are Random Read and Random Writes, not sequential reads or writes. That's what you do when you are loading applications.
Has Anandtech ever reviewed desktop PCIe SATA controller cards?
Review suggestion: include Raid 0 vs. Raid 1 on Areca Raid 1210 controller cards, using Intel SSD vs. WD Raptor HD's.
Maybe include other brands, such as High Point Rocket Raid, 3Ware, Apple's Raid card (for Mac Pro), etc.
How Come Intel Can't Compete The Sequential Performances Of Its Rivals?
Especially the Writing performance.
It seems if Indilinx will match the IO performances of the first Generation of Intel's SSD (It's very close) combining with its better sequential performance and price it will be a much better product.
It seems intel has sacrificed sequential performance for more real world performance. Most people don't spend all day copying lots of single large files.
I'm not sure it's a question of why Intel can't, I think it's a conscious architectural decision. It's something I've been chasing down for months but I can't seem to put my finger on it just yet.
1. Read and write performance is in line with what you'd expect from MLC with 4KB pages and 10 channels, and 1x 2l-plane die/channel: 10x 30MBs = 300MBs sustained read; 10x 7MBs = 70MBs sustained write; read is obviously SATA limited. (I assume they're actually using 2x 32Gb/4GB 34nm stacked die/channel x10 channels for the 80GB drive. Not sure about the 160GB drive.)
2. It appears Intel has gone with more channels and fewer die/channel and maybe smaller page sizes, whereas others have gone with fewer channels and more die/channel and maybe larger page sizes. Intels approach is more expensive, and provides better random write performance at the expense of sequential write performance.
That said, it's curious that the higher write IOPs and reduced write latency, especially on the 160GB drive, don't translate to higher sequential write performance. Maybe a controller limitation? Or maybe the 34nm parts are actually bit slower write than the 50nm parts, but effective throughput is the same due to reduced latency? Or maybe they don't want to risk cannibalizing the E drive market (yet)?
Couple other thoughts... Maybe Intel is being very conservative with these drives and doing read-verify for writes, or intentionally throttling write rates to avoid potential wear problem? I can only imagine the sh*tstorm if drives started failing after a couple years, even if it was a minority of drives/users.
Looking at the SLC E drives, they spec the 32GB drive at 1PB (10^15) lifetime random write, and the 64GB drive at 2PB random write. That yields a write endurance of ~32K cycles/block. MLC drives will be lower, and smaller 34nm geometry will likely be lower still(?) without something else such as additional ECC/spares to compensate. Unfortunately they don't spec the M drives other than a "Minimum of five years of useful life". (Interestingly, both E and M drives are spec'd with the same BER of 1 sector per 10^15 bits read.)
p.s. Sorry the ")" got included in the hyperlink in my previous post; proper URL's are:
I always thought it was because they wanted the M-series to focus on consumer desktop performance. Which means maximizing random performance. It seems to have worked too, as Intel drives are still the SSDs to beat right now.
I don't see any reason why Intel can't reach the same write speeds, I would imagine this 256GB crucial drive is using the same flash, and has similar read/write speeds to the newer samsung's.
It's nice to see ssd drives dropping in price but the capacity still isn't there for the price. I'll stick with my 600gb velociraptors in raid 0. When ssd drives bring out 500 gb and 1tb drives for around $200 or less I'll get a set but I'm not paying over $400 for less than 200gb no.
I think a lot of us are jumping on the Patriot torqx drives because they are hitting a great $/gb (see table). Any info on the new crop of drives would be immensely useful.
Also, Intel is totally missing my price point here, while it is definately a no brainer to get an g2 intel ssd, I dont want a 80gig drive, and 400+ for the 160gig still seems too much for storage.
My guess is that those prices will be fixed for some while....we shall see.
Still don't get SSDs do you? They're not for storage only but for everyday performance. If you do CAD work or movie editing then money in memory+CPU is well spent but not so much for everyday computing.
Intel is missing MY price point by not offering a 40GB model! WinXP and my collection of programs don't exceed 10GB, which leaves 30GB for 5-8 games. 40GB isn't by any means a small amount of flash for a single unit, when you consider that USB drives have been coming in 16MB-4GB for years. I'm not willing to blow $225, but I might spend $115.
On another point, let's look at Intel's price margin. $20 8GB USB drives are everywhere, and they are at least 50nm tech. Buying 10 for $200 would give 80GB, and that's with a loss of efficiency with 10 form factors and controllers. For at least the same price margin, Intel could sell their 34nm 80GB drive for $100.
OSX Trim?
RAID Trim?
If we update to the new firmware will we need to wipe the drive?
Can we run a manual tool to TRIM XP installs?(It'd be good to use with a scheduled task).
I'd like to see benchmarks that target user types.
Workstation, grandparents, gamer, database, web server, etc.
Also, what happens to the SSD industry when only one player has the best tech and lowest prices? Does OCZ kill off the Vertex line?
Can you please benchmark a big RAM drive so we can see what the theoretical best performance is on your benchmarks? It'd be nice to see a best case scenario at the bottom of your charts.
I'm still trying to figure out what Apple's intentions are for TRIM, I have nothing to report there yet. And as far as I know, there's no support for TRIM over a RAIDed volume.
I don't have the TRIM firmware in hand but if it's anything like Intel's previous firmware update, you won't lose your data (it's still always a good idea to back up though).
The performance enhancer tool will work under XP and Vista, I'm guessing it's a manual TRIM utility but I'm waiting to confirm that.
I'm working on individual app benchmarks now, they are just tough to make.
I think OCZ has adopted pretty well thus far, I'd expect that the market will simply require that these drives get even cheaper. I'd look to what OCZ did with its Agility line as an example. Also keep in mind that the conventional HDD makers haven't seriously entered the race yet; once WD and Seagate really get involved then we might see some more competition. Those guys have been working on drive controller technologies for far longer than anyone else, they could be a wild card in this whole thing.
Big RAM drives are interesting but purely from an academic standpoint today. I do want to look at them, but I've gotta get a lot of this other stuff off of my plate first :)
65mW vs 75mW? Big deal. The difference in power between the two devices is 0.01W. Intel is correct. There should be no discernable difference in battery life. For comparison, a single core Intel Atom uses 4W TDP.
The price of two 80gb X25-M drives is only $10 more than the price of the 160gb unit.
Would it make sense to get two smaller drives in raid-0 vs. a single 160gb drive?
There is RAID and there is RAID... If you want to compare RAID performance it would be better not to use wannabe semi-software solutions (Intel ICHxxR and similar). For RAID to do what it is really supposed to do, one needs to look for ~$300+ raid controllers. Ofc, even semi-software solutions give you a boost, but it's better not compared to pure hardware ones (usualy hardware ones have also SODIMM DDR2 memory working as cache either in form of a add-in slot or just chips on the board).
Any way, RAIDing X25M seems to me a bit contraproductive... It won't decrease random write latency much (well, it does if you enable your raid controller to use it's RAM to cache writes as well, but it is not a RAID feature, it is controller feature), just increases max read/write transfer speed - useless unless you are server or work with uncompressed video.
So either you RAID1 your SSDs to make backup in case _only one fails_ or make RAID5(e) for increased max transfer speed & _only one fails_ backup.
Just my opinion, not claiming it has to be 100% correct in all cases.
It would be interesting if you could make some day an article about RAID 0 and RAID 1 and its usefulness on desktops (with benchmarks baking up the data). It's very difficult to find useful information about that on the internet (that is not 5 years old).
Which begs the question, do you get better performance with two drives in RAID 0 or with one drive using TRIM? I guess we won't know until you get your drives for testing.
Have you asked them for more than just one drive so you can test the functionality of RAID? Thanks.
Well that cements this as my next upgrade. I think I'll order one in Oct when I upgrade my system to W7, might even manage to get the price down a little more by then. Go competition!
Why? Albeit expensive I think there's going to be stronger demand on higher capacities than continuing to improve performance. If they seriously want these to replace traditional HDDs they've gotta ramp up capacity.
"Load power hasn't changed between the drives, they are both at 150mW. Idle power went up by 10mW thanks to the new controller. These are "typical" values while running MobileMark 2007, so I'm not sure if things change at all if you're looking at real world workloads. Intel claims that the new drive doesn't deliver worse battery life than the old one, but I'll have to verify."
See, if they follow the logic of "Work faster to get to idle", the substantially increased random write performance will more than make up the 10mW increased idle power.
I'm ordering two 80GBs as soon as they hit the shelves!
The news about 50nm models not getting the TRIM support is rather disheartening, though. I'm not sure why Intel would basically screw those that forked over the HUGE cash for the early models.
Can someone please comment on TRIM support with the older X25-M. As I understand it, because of the architecture of the X25M and it's far superior random write IOPS, TRIM doesn't make a lot of difference. I couldn't find all the links, but go here and scroll to the bottom:
http://www.pcper.com/#NewsID-7488">http://www.pcper.com/#NewsID-7488
Can the author comment on this please? After Intel provides new firmware that fixed the major slowdown on drives over time, what type of impact will be caused by not having TRIM support?
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
73 Comments
Back to Article
Bolas - Monday, November 23, 2009 - link
MSRP turns out to be way too optimisitc.Drive NAND Capacity Cost per GB Price
Intel X25-M (34nm) 80GB $2.81 $225
Intel X25-M (34nm) 160GB $2.75 $440
Those prices still aren't available four months after launch. Would have been nice, though.
Pandamonium - Wednesday, January 20, 2010 - link
X18-M still nowhere to be found. This is really frustrating.sotoa - Thursday, July 23, 2009 - link
SSD's get me excited for some reason. They're getting faster, evolving, getting cheaper, are mysterious, yet Anandtech peels away the mystery.I can't get enough.
I can't wait to get my Win7 pro, and SSD in October. I'll decide in October which SSD, but Intel just made it sweeter.
futrtrubl - Wednesday, July 22, 2009 - link
One interesting but ultimately irrelevant derived statistic from the G rating of this drive...It can be accelerated to 7.35km/s over that 0.5 seconds.... discounting drag that's the delta-v required for low earth orbit....
Unfortunately that would require a canon/rail over 3km long ;']
GTVic - Wednesday, July 22, 2009 - link
I would like to kill (read: eliminate from existence) the tech writer, probably from HardOCP (read: guy who likes to review power supplies and use the non-word "immersive" way too much), who started using this "read" (read: read) terminology to explain what he meant.Please, why not just write what you mean.
Pastuch - Wednesday, July 22, 2009 - link
I have an ADFD generation Raptor and I'm thinking about getting the 80gb 34nm Intel SSD.
How would I install the 2.5 inch drive into my Antec P160 or Antec Gamer 300 cases? Do the drives come with 3.5 inch adapters?
rcpratt - Wednesday, July 22, 2009 - link
I haven't ever really tried to purchase brand-spanking new hardware before, so one question - how long does it usually take for newegg to start selling new hardware?rhys216 - Wednesday, July 22, 2009 - link
It is impossible for car makers to do this!Stop comparing APPLES to ORANGES.
People have paid through the nose for Intel drives and now Intel have left them bloody.
It's a simple fix that Intel are CHOOSING to do, and yes that is why people are feeling screwed.
I bet Intel would have to steal your wife and kill your dog before you said a bad word about them!!!
F**K Intel
dangerz - Wednesday, July 22, 2009 - link
Why can't we forget SATA and use PCI-E?xyxer - Tuesday, July 21, 2009 - link
Crap I bought 3 50nms last month?! Intel wouldn't be so stupid to screws its customers would it ?michal1980 - Wednesday, July 22, 2009 - link
hows it screwing you?when car makers add new features to new model years are they screwing the prievous model year buyers?
iwodo - Tuesday, July 21, 2009 - link
I forgot to mention, doesn't 150mw seems very low? Other site has been reporting other SSD using up to 4W when active. Which means it aren't saving much energy compare to 2.5" HDD.May be anand could do a power table comparison when you ever decide to do an SSD roundup.
iwodo - Tuesday, July 21, 2009 - link
If the Vertex gets anywhere between 20 - 30% price cut it will be a very attractive price point for everyone.However at the current rate i fail to see why i would choose anything over Intel 34nm offering.
Anand do you know, what processing technology is being used for Intel Controller chip?
It is rather disappointing that even Intel upcoming chipset wont have SATA 3.0. The current 2.0 is seriously holding back SSD RAW performance.
It looks like 2010 will finally be the year of SSD.
blyndy - Tuesday, July 21, 2009 - link
There have been a lot of new SSD releases and updates in the past six months and even in the past few weeks.It's time for an SSD roundup.
VooDooAddict - Tuesday, July 21, 2009 - link
I was looking at the Intel 80GB over the weekend for an i7 build. This price drop is great news.BenJackson - Tuesday, July 21, 2009 - link
If I understand Trim correctly from your previous article http://www.anandtech.com/storage/showdoc.aspx?i=35...">http://www.anandtech.com/storage/showdoc.aspx?i=35..., it will delete and re-write blocks after every file delete operation to clean the LBA page, whether this will be helpful or not.Considering consumer drives are MLC-based, do you have any figures on how the Trim command may affect the life-span of these drives?
vol7ron - Tuesday, July 21, 2009 - link
Could you please compare to the new Corsair coming out?Most notably the x64 or x128
http://www.engadget.com/2009/07/16/corsair-cranks-...">http://www.engadget.com/2009/07/16/cors...-extreme...
http://hothardware.com/News/Corsair-Launches-New-E...">http://hothardware.com/News/Corsair-Launches-New-E...
I'm not sure if you could get your hands on one, but it'd be nice to see.
Thank you,
vol7ron
Endoplasmic - Tuesday, July 21, 2009 - link
"TRIM Support: Not For 50nm Drives"I was hoping you could point to a link that states this from Intel. My x25-m is fairly new (less than a week) and I'd like to make sure that if I was to return it that it was because TRIM isn't supported on this drive.
The read/write stuff I can live without, but extending the life of the drive is what I'm down with.
Thanks.
snoooze - Tuesday, July 21, 2009 - link
Since SSD lack rotating parts and are pretty much unsensitive to dust I can't really see the reason to continue with an enclosure and the accompanying cable harness that is the case with a 2.5"/SATA-drive. And how long will it take before even SATA 6Gbit is saturated performance wise, 2-3 years?I would really like to see a PCI-express version of this and other new cheap performance oriented SSD.
Would be interesting to here your word on this Anand.
Best regards
snoooze - Tuesday, July 21, 2009 - link
hear*dirt2901 - Tuesday, July 21, 2009 - link
The one thing I was wondering about SSD's was how good they would be for the indexes on a newgroup reader> Specifically using Agent version 5 and putting it on an SSD by it self and letting the reader grab over 40G of data. Then when it does the purge and compact on the databases I wondered how it would stack up with a typical hard drive. Does anyone have any ideas if this would be a good use or not??Thanks!
snorp - Tuesday, July 21, 2009 - link
Anybody know when we'll be able to buy these? I want.NKnight - Tuesday, July 21, 2009 - link
Since Intel doesn't appear to be introducing new SLC Extreme drives in the near future, any word of a firmware update for the 50nm 25-E drives to include TRIM support when Win7 is released in October?deanx0r - Tuesday, July 21, 2009 - link
I didn't see it mentioned in the article, but I presume this is still a SATA 2 device?rcpratt - Tuesday, July 21, 2009 - link
Yes.jhoff80 - Tuesday, July 21, 2009 - link
My (limited) understanding is that it's not quite as necessary for the TRIM command for an SLC drive. However, it seems to be that it couldn't hurt, right?With that being the case, any idea if Intel will be working that into the firmware for the x25-E, since they don't plan to introduce a 34nm model?
yyrkoon - Tuesday, July 21, 2009 - link
"The new controller is Halogen-free (the old one wasn’t) so Apple could theoretically use the new drives in their systems without being un-green."Yeah, because it is very important that Apple "feels" green, after their Mac pro burns off ~.3KW's worth of power. Theirs are not the only systems that use "too much power" though.
One of these days, OEMs will get a clue, and perhaps then push the industry into using truly "green" components. It is great that we now have RoHS ( although solder is a mixed bag if it does not bond right, or well ), and low power usage components such as the SSD. Maybe some day CPU/GPU manufactures will get a clue ? Sure, *someone* needs that 1kW space heater of a PC. Now we need something that can do most tasks without forcing the user to buy a laptop/netbook, and using enough energy to power 2-3 mid sized refrigerators.
mikesown - Tuesday, July 21, 2009 - link
Any word on whether the drives are ONFI 1.0 or 2.1? I'm assuming, based on the speeds, that it's (unfortunately, and strangely) the former. I'm disappointed that Intel didn't use the new Micron 34nm ONFI 2.1 NAND.has407 - Tuesday, July 21, 2009 - link
ONFI 1.0; see my previous post.ilkhan - Tuesday, July 21, 2009 - link
I think Intel has an independent RAM source, its not from Micron. AFAIK anyway.mikesown - Tuesday, July 21, 2009 - link
I'm not referring to the RAM cache - I'm referring to the NAND used for storage. Intel partnered with Micron to form IM flash technologies, and Intel uses the NAND from IM in their SSDs.fredsky - Tuesday, July 21, 2009 - link
hello Intel !nice update BUT the most important thing is the WRITE speed.
why in hell are you limited (artificially) to 70MB/s ? go 200MB/s please!
erple2 - Tuesday, July 21, 2009 - link
No, the most important thing for drives of this size are Random Read and Random Writes, not sequential reads or writes. That's what you do when you are loading applications.JohanAnandtech - Wednesday, July 22, 2009 - link
True. People forget how many times the OS has to access and write to the master file table. Those are all small block accesses.Also, desktops read a lot more than they write. So it is hard to imagine that anyone will be experiencing the write speed of 70 MB/s as a bottleneck.
rcpratt - Tuesday, July 21, 2009 - link
Hurry up, newegg :(vailr - Tuesday, July 21, 2009 - link
Has Anandtech ever reviewed desktop PCIe SATA controller cards?Review suggestion: include Raid 0 vs. Raid 1 on Areca Raid 1210 controller cards, using Intel SSD vs. WD Raptor HD's.
Maybe include other brands, such as High Point Rocket Raid, 3Ware, Apple's Raid card (for Mac Pro), etc.
Drazick - Tuesday, July 21, 2009 - link
How Come Intel Can't Compete The Sequential Performances Of Its Rivals?Especially the Writing performance.
It seems if Indilinx will match the IO performances of the first Generation of Intel's SSD (It's very close) combining with its better sequential performance and price it will be a much better product.
Sunday Ironfoot - Tuesday, July 21, 2009 - link
It seems intel has sacrificed sequential performance for more real world performance. Most people don't spend all day copying lots of single large files.jimhsu - Wednesday, July 22, 2009 - link
PCPer believes that the waveform for seq writes is clipped at 70 MB/s on purpose: http://www.pcper.com/article.php?aid=691&type=...">http://www.pcper.com/article.php?aid=691&type=...I would say most likely product differentiation. And maybe the fact that the X25 doesn't cache writes unlike Indilinx makes a difference.
Anand Lal Shimpi - Tuesday, July 21, 2009 - link
I'm not sure it's a question of why Intel can't, I think it's a conscious architectural decision. It's something I've been chasing down for months but I can't seem to put my finger on it just yet.Take care,
Anand
has407 - Tuesday, July 21, 2009 - link
From the Intel product page (http://www.intel.com/design/flash/nand/mainstream/...">http://www.intel.com/design/flash/nand/mainstream/...:"10 Parallel Channel Architecture with 34nm MLC ONFI 1.0 NAND (34nm)"
[NOTE: same for 50nm devices]
Based on numbers from Intel/ONFI presentations (http://onfi.org/presentations/)">http://onfi.org/presentations/)...
1. Read and write performance is in line with what you'd expect from MLC with 4KB pages and 10 channels, and 1x 2l-plane die/channel: 10x 30MBs = 300MBs sustained read; 10x 7MBs = 70MBs sustained write; read is obviously SATA limited. (I assume they're actually using 2x 32Gb/4GB 34nm stacked die/channel x10 channels for the 80GB drive. Not sure about the 160GB drive.)
2. It appears Intel has gone with more channels and fewer die/channel and maybe smaller page sizes, whereas others have gone with fewer channels and more die/channel and maybe larger page sizes. Intels approach is more expensive, and provides better random write performance at the expense of sequential write performance.
That said, it's curious that the higher write IOPs and reduced write latency, especially on the 160GB drive, don't translate to higher sequential write performance. Maybe a controller limitation? Or maybe the 34nm parts are actually bit slower write than the 50nm parts, but effective throughput is the same due to reduced latency? Or maybe they don't want to risk cannibalizing the E drive market (yet)?
has407 - Tuesday, July 21, 2009 - link
Couple other thoughts... Maybe Intel is being very conservative with these drives and doing read-verify for writes, or intentionally throttling write rates to avoid potential wear problem? I can only imagine the sh*tstorm if drives started failing after a couple years, even if it was a minority of drives/users.Looking at the SLC E drives, they spec the 32GB drive at 1PB (10^15) lifetime random write, and the 64GB drive at 2PB random write. That yields a write endurance of ~32K cycles/block. MLC drives will be lower, and smaller 34nm geometry will likely be lower still(?) without something else such as additional ECC/spares to compensate. Unfortunately they don't spec the M drives other than a "Minimum of five years of useful life". (Interestingly, both E and M drives are spec'd with the same BER of 1 sector per 10^15 bits read.)
p.s. Sorry the ")" got included in the hyperlink in my previous post; proper URL's are:
http://www.intel.com/design/flash/nand/mainstream/...">http://www.intel.com/design/flash/nand/mainstream/...
http://onfi.org/presentations/">http://onfi.org/presentations/
Kurotetsu - Tuesday, July 21, 2009 - link
I always thought it was because they wanted the M-series to focus on consumer desktop performance. Which means maximizing random performance. It seems to have worked too, as Intel drives are still the SSDs to beat right now.stevty2889 - Tuesday, July 21, 2009 - link
I don't see any reason why Intel can't reach the same write speeds, I would imagine this 256GB crucial drive is using the same flash, and has similar read/write speeds to the newer samsung's.http://www.crucial.com/store/partspecs.aspx?IMODUL...">http://www.crucial.com/store/partspecs.aspx?IMODUL...
NeBlackCat - Tuesday, July 21, 2009 - link
I've seen speculation that the new drives would have built-in encryption, no sign of that?Anand Lal Shimpi - Tuesday, July 21, 2009 - link
No encryption for this controller :)Take care,
Anand
Soldier1969 - Tuesday, July 21, 2009 - link
It's nice to see ssd drives dropping in price but the capacity still isn't there for the price. I'll stick with my 600gb velociraptors in raid 0. When ssd drives bring out 500 gb and 1tb drives for around $200 or less I'll get a set but I'm not paying over $400 for less than 200gb no.benrico - Tuesday, July 21, 2009 - link
I think a lot of us are jumping on the Patriot torqx drives because they are hitting a great $/gb (see table). Any info on the new crop of drives would be immensely useful.Also, Intel is totally missing my price point here, while it is definately a no brainer to get an g2 intel ssd, I dont want a 80gig drive, and 400+ for the 160gig still seems too much for storage.
My guess is that those prices will be fixed for some while....we shall see.
semo - Friday, July 24, 2009 - link
Still don't get SSDs do you? They're not for storage only but for everyday performance. If you do CAD work or movie editing then money in memory+CPU is well spent but not so much for everyday computing.HDDs are for storage people!
AnnonymousCoward - Friday, July 24, 2009 - link
Intel is missing MY price point by not offering a 40GB model! WinXP and my collection of programs don't exceed 10GB, which leaves 30GB for 5-8 games. 40GB isn't by any means a small amount of flash for a single unit, when you consider that USB drives have been coming in 16MB-4GB for years. I'm not willing to blow $225, but I might spend $115.On another point, let's look at Intel's price margin. $20 8GB USB drives are everywhere, and they are at least 50nm tech. Buying 10 for $200 would give 80GB, and that's with a loss of efficiency with 10 form factors and controllers. For at least the same price margin, Intel could sell their 34nm 80GB drive for $100.
mckirkus - Tuesday, July 21, 2009 - link
Anand, for a full review, some ideas.OSX Trim?
RAID Trim?
If we update to the new firmware will we need to wipe the drive?
Can we run a manual tool to TRIM XP installs?(It'd be good to use with a scheduled task).
I'd like to see benchmarks that target user types.
Workstation, grandparents, gamer, database, web server, etc.
Also, what happens to the SSD industry when only one player has the best tech and lowest prices? Does OCZ kill off the Vertex line?
Can you please benchmark a big RAM drive so we can see what the theoretical best performance is on your benchmarks? It'd be nice to see a best case scenario at the bottom of your charts.
Anand Lal Shimpi - Tuesday, July 21, 2009 - link
I'm still trying to figure out what Apple's intentions are for TRIM, I have nothing to report there yet. And as far as I know, there's no support for TRIM over a RAIDed volume.I don't have the TRIM firmware in hand but if it's anything like Intel's previous firmware update, you won't lose your data (it's still always a good idea to back up though).
The performance enhancer tool will work under XP and Vista, I'm guessing it's a manual TRIM utility but I'm waiting to confirm that.
I'm working on individual app benchmarks now, they are just tough to make.
I think OCZ has adopted pretty well thus far, I'd expect that the market will simply require that these drives get even cheaper. I'd look to what OCZ did with its Agility line as an example. Also keep in mind that the conventional HDD makers haven't seriously entered the race yet; once WD and Seagate really get involved then we might see some more competition. Those guys have been working on drive controller technologies for far longer than anyone else, they could be a wild card in this whole thing.
Big RAM drives are interesting but purely from an academic standpoint today. I do want to look at them, but I've gotta get a lot of this other stuff off of my plate first :)
Take care,
Anand
lotharamious - Tuesday, July 21, 2009 - link
65mW vs 75mW? Big deal. The difference in power between the two devices is 0.01W. Intel is correct. There should be no discernable difference in battery life. For comparison, a single core Intel Atom uses 4W TDP.geofelt - Tuesday, July 21, 2009 - link
The price of two 80gb X25-M drives is only $10 more than the price of the 160gb unit.Would it make sense to get two smaller drives in raid-0 vs. a single 160gb drive?
Passing by - Tuesday, July 21, 2009 - link
Not an option for most notebooks.YGDRASSIL - Tuesday, July 21, 2009 - link
Yes it would. Googling along you will find some reviews. But you would want seperate sata channels for each drive.Anand Lal Shimpi - Tuesday, July 21, 2009 - link
Two things:1) I'm not sure if TRIM will work on drives in RAID.
2) The performance benefit from RAID isn't as great in the real world as you'd expect.
Take care,
Anand
HollyDOL - Wednesday, July 22, 2009 - link
There is RAID and there is RAID... If you want to compare RAID performance it would be better not to use wannabe semi-software solutions (Intel ICHxxR and similar). For RAID to do what it is really supposed to do, one needs to look for ~$300+ raid controllers. Ofc, even semi-software solutions give you a boost, but it's better not compared to pure hardware ones (usualy hardware ones have also SODIMM DDR2 memory working as cache either in form of a add-in slot or just chips on the board).Any way, RAIDing X25M seems to me a bit contraproductive... It won't decrease random write latency much (well, it does if you enable your raid controller to use it's RAM to cache writes as well, but it is not a RAID feature, it is controller feature), just increases max read/write transfer speed - useless unless you are server or work with uncompressed video.
So either you RAID1 your SSDs to make backup in case _only one fails_ or make RAID5(e) for increased max transfer speed & _only one fails_ backup.
Just my opinion, not claiming it has to be 100% correct in all cases.
Mr Perfect - Tuesday, July 21, 2009 - link
3) If one drive fails, the other drive holds useless data.I always kind of cringe when people make RAID 0 arrays. Why no love for RAID that's actually redundant?
pmonti80 - Tuesday, July 21, 2009 - link
It would be interesting if you could make some day an article about RAID 0 and RAID 1 and its usefulness on desktops (with benchmarks baking up the data). It's very difficult to find useful information about that on the internet (that is not 5 years old).RaistlinZ - Tuesday, July 21, 2009 - link
Which begs the question, do you get better performance with two drives in RAID 0 or with one drive using TRIM? I guess we won't know until you get your drives for testing.Have you asked them for more than just one drive so you can test the functionality of RAID? Thanks.
rcpratt - Tuesday, July 21, 2009 - link
I am. The only reason not to would be the 160GB models slightly higher 4k random writes.djc208 - Tuesday, July 21, 2009 - link
Well that cements this as my next upgrade. I think I'll order one in Oct when I upgrade my system to W7, might even manage to get the price down a little more by then. Go competition!jwilliams4200 - Tuesday, July 21, 2009 - link
There were rumors of 320GB drives. This seems logical given the doubling of Flash chip capacity.Why no mention of 320GB drives?
Anand Lal Shimpi - Tuesday, July 21, 2009 - link
I don't believe we'll see 320GB drives anytime soon.Take care,
Anand
Denithor - Wednesday, July 22, 2009 - link
Why? Albeit expensive I think there's going to be stronger demand on higher capacities than continuing to improve performance. If they seriously want these to replace traditional HDDs they've gotta ramp up capacity.IntelUser2000 - Tuesday, July 21, 2009 - link
"Load power hasn't changed between the drives, they are both at 150mW. Idle power went up by 10mW thanks to the new controller. These are "typical" values while running MobileMark 2007, so I'm not sure if things change at all if you're looking at real world workloads. Intel claims that the new drive doesn't deliver worse battery life than the old one, but I'll have to verify."See, if they follow the logic of "Work faster to get to idle", the substantially increased random write performance will more than make up the 10mW increased idle power.
melgross - Tuesday, July 21, 2009 - link
Is it known whether Apple will implement trim in 10.6?rcpratt - Tuesday, July 21, 2009 - link
I'm ordering two 80GBs as soon as they hit the shelves!The news about 50nm models not getting the TRIM support is rather disheartening, though. I'm not sure why Intel would basically screw those that forked over the HUGE cash for the early models.
winterspan - Tuesday, July 21, 2009 - link
Can someone please comment on TRIM support with the older X25-M. As I understand it, because of the architecture of the X25M and it's far superior random write IOPS, TRIM doesn't make a lot of difference. I couldn't find all the links, but go here and scroll to the bottom:http://www.pcper.com/#NewsID-7488">http://www.pcper.com/#NewsID-7488
Can the author comment on this please? After Intel provides new firmware that fixed the major slowdown on drives over time, what type of impact will be caused by not having TRIM support?
TemjinGold - Tuesday, July 21, 2009 - link
So they'd buy new drives, of course... ;)rcpratt - Tuesday, July 21, 2009 - link
Well, obviously. Just seems like something Intel doesn't usually do.icrf - Tuesday, July 21, 2009 - link
It might be b/c TRIM postpones the "used" state of a drive, and many of the existing units may have already hit this used state.