Ads
Ads
All about Storage Solutions
 Latest Storage Solutions articles
Perfect Consumer Backup With Seagate And Rebit

Perfect Consumer Backup With Seagate And Rebit
Seagate’s Replica is a portable hard drive with automatic backup, versioning, and disaster recovery features to make backup a no-brainer. The product works so well that it deserves our rare Best of Tom’s Award. Read More

  • Thecus N8800 NAS/SAN: Quick-Look
    After recently running out of room on my Thecus N5200, I again turned to the company for something larger. What I found won't work for most folks (unless you have a rack in your garage, that is). But it's reasonably-priced and plenty fast for most SMBs. Read More
All Storage Solutions articles

Newsletters


  • Ask your question about IT issues
  • Post

Partners

The Games selection

violent : Interactive Buddy Unwind on your interactive buddy: Do anything you want to him, it will earn you money, and you can buy other stuff to torture him with.
violent : More Mindless Violence Basic shooting game, but still so powerful! Use the mouse to take aim and shoot at the little beasties before they get to you. Use Space to reload....
Ads

Sponsored links

Intel Confirms, Replicates SSD Firmware Bug

Next news
2:41 PM - November 6, 2009 by Marcus Yam

Intel now working on a fix.

Late last month, Intel released a firmware update for its second-generation 34nm X25-M solid state drive line that adds support for the TRIM command--which keeps SSDs operating at a higher level of performance.

Sadly, instead of getting performance benefits, some users running Windows 7 64-bit found that the firmware bricked their precious SSDs--definitely not the effect that anyone wanted.

Intel pulled the firmware following the failure reports. The chip giant today confirmed that it's been able to replicate the fatal flaw that has plagued users.

According to the Register, Alan Frost of Intel's NAND Solutions Group wrote: "Intel has replicated the issue on 34nm SSDs - X25-M - and is working on a fix."

It seems that Intel is asking for outside help to resolve this issue, as Frost added, "Intel is seeking direct feedback on this issue from members of the [Intel Support Community]... asking them to send their drives directly to Intel to expedite the analysis of the issues. This action will enable us to more quickly generate a resolution for this issue."

Source : Tom's Hardware US

Talkback
Add your comment
doomtomb 11/06/2009 9:06 PM
Hide
-20+

Intel has been slacking recently. Exorbitant prices and things that don't work, disregarding USB 3.0 as unimportant... wow... Intel needs to get slapped in the face right about now.

D_Kuhn 11/06/2009 9:06 PM
Hide
-5+

That would REALLY SUCK. Getting all wound up to go even faster with a hot new firmware release on an already fast (and $$$) SSD... and to have it brick the machine on reboot.

Ouch!

Onyx2291 11/06/2009 9:07 PM
Hide
-2+

Oh man that sucks to have that bricked. Unfortunate for an update like that would cause that. I don't have one but hope a fix comes up soon.

El_Capitan 11/06/2009 9:09 PM
Hide
-0+

I've been commenting since forever that Intel's X25-M is an overpriced piece of dud that for some reason OEM's only use for SSD's. Crucial's M225 series and Corsair's P series are the ones to get.

Too bad every distributor out there is rising their SSD prices for Black Friday (day after Thanksgiving), so that when they go on sale, they get back to the already pricey $350 pricetag. No illusion, check out all the prices, SSD's for 128GB are hitting $450, when just 7 months ago they were at $280.

ssalim 11/06/2009 9:13 PM
Show
lubitz_420 11/06/2009 9:16 PM
Hide
-4+

good to see they don't just try to sweep the problem under the rug by offering a replacement to those who lost their drive functionality and then just letting it trail off into oblivion +1 to solving the problem rather than hiding it

spectrewind 11/06/2009 9:33 PM
Hide
-2+

ssalim :
Since when did "bricked" become a word...?



See the word "Meh" for how this happens. ;o)

ctbaars 11/06/2009 9:34 PM
Show
El_Capitan 11/06/2009 9:36 PM
Hide
--3+

I like how I get a thumbs every time I comment on the Intel X250M, but on all my other comments always gets a thumb's up. There's some Intel fanboy's in this thread. :)

El_Capitan 11/06/2009 9:40 PM
Hide
-3+

ctbaars :
@ El_Capitan. I'm having a hard time finding a post by you that says, "Intel's X25-M is an overpriced piece of dud that for some reason OEM's only use for SSD's"


I guess I found one Intel fanboi. I don't post threads like that. I only comment on the news sections. They're probably hidden because the Intel fanboi's like giving my comments the thumbs down.

You'll find me repeatedly telling people that the Prices per Storage and Performance is vastly inferior (besides throughput) to that of the once-cheaper Crucial CT128M225 and Corsair P128. There was even an article a few months back noting the comparisons (but on 256GB versions).

El_Capitan 11/06/2009 9:45 PM
Hide
-2+

ctbaars :
@ El_Capitan. I'm having a hard time finding a post by you that says, "Intel's X25-M is an overpriced piece of dud that for some reason OEM's only use for SSD's"



Intel to Ship Value-Based X25-X SSD in Q4
10-16-2009 at 09:25:58 PM
Besides for throughput, Corsair's CT128M225 and Crucial's P128 are much better overall. Plus, I'm starting to get rather annoyed. What's with this new business model over the last few years? I remember when the latest and best model got the high price tag, then dropped in price as a newer and better model came out. Nowadays, the latest model price stays the same (or in SSD's cases, goes up) when a newer but lower performance model comes out.

Intel Pulls X25-M G2 TRIM Firmware Update
10-27-2009 at 11:44:36 PM
For those that put a thumbs down on my last post about the Intel X-35m... haha.

*****

Well, if you fanboi's don't keep giving a thumb's down to my comments, you'd probably have read them.

El_Capitan 11/06/2009 9:49 PM
Hide
--1+

ctbaars :
@ El_Capitan. I'm having a hard time finding a post by you that says, "Intel's X25-M is an overpriced piece of dud that for some reason OEM's only use for SSD's"


Oh yeah, another hidden one.

New Intel 34nm X25-M SSD Firmware Brings Impressive Performance Gains
10-26-2009 at 02:02:40 PM
A new firmware for the Intel X25-M G1 is nice and all, but the SSD still fails price per performance per space against Crucial's M225 and Corsair's P128/P256. All you get with the X25-M G1 is great I/O performance. Write speeds are still pretty low compared to going up to 200MB's W/s.

Intel X25-m = 160GB, $659.00, 250 MB/s Read, 70 MB/s Write
Crucial M225 = 256GB, $675.00, 250 MB/s Read, 200 MB/s Write
Corsair P256 = 256GB, $719.00 (free shipping), 220 MB/s Read, 200 MB/s Write

Prices are from Newegg's retail prices. You can get them cheaper other places and OEM.

*****

gg, no re

anamaniac 11/06/2009 9:49 PM
Hide
-0+

Well, atleast they're fixing it.

ctbaars 11/06/2009 10:09 PM
Hide
--2+

rofl! Thanks, El_Capitan. I guess you're a fan of Corsair's CT128M225 and Crucial's P128's ;)

El_Capitan 11/06/2009 10:20 PM
Hide
-4+

ctbaars :
rofl! Thanks, El_Capitan. I guess you're a fan of Corsair's CT128M225 and Crucial's P128's


I'm a fan of the better products, then put pricing into the equation. Intel, AMD, NVidia, ATI, Crucial, Corsair, Kingston, Western Digital, Seagate, etc...

That's what Tom's Hardware was supposed to be about with their articles, but it's becoming less and less that way.

goodguy713 11/06/2009 10:59 PM
Hide
-0+

well if they can drag usb 3.0 out longer then they stand to make more money .because they can offer it as a massive upgrade. right now there is no real pressure on them to put out faster cpu's or add usb 3.0 because they are on top of the market. AMD is still at best a year behind them. also i would not be surprised if some thing else was holding them back. usb 3.0 would be awesome to have thing is even the ssd drives barely are able to keep up with the maximum read cycles ..

cdillon 11/07/2009 12:11 PM
Hide
-2+

@El_Capitan: You're only looking at Sequential read/write performance. If that's all you really care about then, yes, the Intel drives are over-priced for what they give you. But take a look at random read/write performance and especially performance scaling when the queue-depth rises on the Intel drives. They blow those cheaper drives completely out of the water on random or heavily concurrent workloads where sequential rates don't matter nearly as much. The Intel drives are more balanced instead of trying to focus too much on sequential transfer rates. The whole wide world of computing isn't made just for average desktop users, there ARE other needs out there.

rippleyhakd 11/07/2009 2:10 AM
Hide
-0+

I dont know what scares me more.. A backpedling internet guy, or a forum stalker that actually has remarks from 20+ days ago, at his finger tips..

digitalrazoe 11/07/2009 2:39 AM
Hide
-0+

hmm... looks like I will keep using my "spinners" for a long time to come. ( $99 dollar 1.5TB 60MB avg bandwidth harddrive failure or a 600+ 160gig 200MB SSD failure ?) yup.. I will take the slow lane for now ( you spin me right round baby right round..)

El_Capitan 11/07/2009 2:59 AM
Hide
--2+

rippleyhakd :
I dont know what scares me more.. A backpedling internet guy, or a forum stalker that actually has remarks from 20+ days ago, at his finger tips..


Yeah, like clicking on Forums and See All threads is a difficult thing to do... I'd go with the backpedaling internet girl.

El_Capitan 11/07/2009 3:18 AM
Hide
--3+

cdillon :
@El_Capitan: You're only looking at Sequential read/write performance. If that's all you really care about then, yes, the Intel drives are over-priced for what they give you. But take a look at random read/write performance and especially performance scaling when the queue-depth rises on the Intel drives. They blow those cheaper drives completely out of the water on random or heavily concurrent workloads where sequential rates don't matter nearly as much. The Intel drives are more balanced instead of trying to focus too much on sequential transfer rates. The whole wide world of computing isn't made just for average desktop users, there ARE other needs out there.


Of course, but if you look at all the desktops and laptops that can be customized in the market for average desktop users, they go with only the Intel SSD's. Any Enterprise user out there for servers will still go with SCSI drives than the Intel SSD. They won't offer the Intel SSD for servers, and for good reason.

zerapio 11/07/2009 4:03 AM
Hide
-0+

El_Capitan :
Of course, but if you look at all the desktops and laptops that can be customized in the market for average desktop users, they go with only the Intel SSD's. Any Enterprise user out there for servers will still go with SCSI drives than the Intel SSD. They won't offer the Intel SSD for servers, and for good reason.



So you're saying that businesses care more about sequential performance?

Anonymous 11/07/2009 5:04 AM
Hide
-0+

El_Capitan, I'm not sure why an enterprise user wouldn't want to go with an Intel SSD:
http://www.anandtech.com/storage/showdoc.aspx?i=3631
http://it.anandtech.com/IT/showdoc.aspx?i=3532&p=11

backIn5 11/07/2009 6:33 AM
Hide
-0+

This reminds me of the Seagate firmware fiasco, a year ago. Although, that was caused by factory test misapplication.

Let's hope Intel handles this problem better than Seagate did.

El_Capitan 11/07/2009 6:34 AM
Hide
-0+

asdf123456789 :
El_Capitan, I'm not sure why an enterprise user wouldn't want to go with an Intel SSD:http://www.anandtech.com/storage/showdoc.aspx?i=3631http://it.anandtech.com/IT/showdoc.aspx?i=3532&p=11


Can't argue that. Man, that's a great article.

fourfives 11/07/2009 2:42 PM
Hide
-0+

I would think most businesses require massive random reads/writes and IOPS not seq reads. This drive does that very very well. Now if only trim ran on raid, within an iscsi SAN...

Dkz 11/07/2009 2:55 PM
Hide
-0+

Too early to not have problems, this tech it's just too fresh and too fast it needs to have something wrong.. Moors law right?
Maybe Intel it's being too aggressive selling out the drives without a proper testing. Even if i could afford those drives right now i wouldn't buy one not just now, maybe some other brand with less issues reports ;x

spoofedpacket 11/07/2009 3:32 PM
Hide
-0+

El_Capitan :
Intel X25-m = 160GB, $659.00, 250 MB/s Read, 70 MB/s



I know the channel price for this item is $432.00 on any given day. The reseller I picked up from gave a fair price of $462.00. Channel prices on the other two you listed are considerably more and not just a high demand markup by Newegg, but the actual common price.

As for Intel drives and OEMs, someone needs to give some citations. Just about every major line I've seen is using SamSung in their current offerings.

wildwell 11/07/2009 10:27 PM
Hide
-0+

ssalim :
Since when did "bricked" become a word...?


Being a Regular at Tom's, I'm surprised you've never seen, "Bricked."

El_Capitan 11/08/2009 9:25 PM
Hide
-0+

Reading user reviews of the product at various places (Newegg, Amazon, etc.), the users are experiencing: freezes, stuttering, lock-ups, and failure after several months of usage. While the professional reviews of the Intel X25-M regard it highly as the best SSD out there, and now this firmware bug, I think every fan of this SSD got played. I'll stick with my reliable SSD's. No point in shelling out $600 for a broken product. gg Intel, well played.

master exon 11/08/2009 10:02 PM
Hide
-0+

ssalim :
Since when did "bricked" become a word...?


It's been a technical term in the electronic world for a very long time. Glad you came out from under your rock.


Sponsored links

Related articles

  • Throughput Testing Do you remember read performance crashing to 42 MB/s with the old firmware? Intel’s engineers did a great job in optimizing the firmware, as the read throughput didn’t decrease significantly any more with the new 8820 firmware. The minimum throughput after the first h2benchw/IOMeter cycle was 109 MB/s, but after just the second cycle the X25-M was able to adjust and maintain a minimum sequential read throughput of 200 MB/s. Although the firmware update cannot prevent minimum write throughput decreasing to only a few megabytes per second, the maximum results stay at a constant 80 MB/s, while the average numbers even keep increasing. Clearly, the firmware adjusts to the workload much more efficiently. I/O Testing There still is a little performance drop, but the 8820 firmware manages to buffer the drop significantly. A decrease from 3,300 to 3,100 I/O operations per second for typical database transactions is certainly acceptable. The extreme performance differences of the 8160 firmware are gone with version 8820, but the fileserver performance keeps increasing. Web server performance doesn’t suffer from the heavily changing torture testing with the new firmware, as the X25-M reaches 11,400 I/O operations per second in the Web server test at all times. Workstation performance, which is based on small to medium block sizes, is different now. Although the first run results in less I/O performance than was the case on the old firmware, the following runs actually are much better. Overall, the new 8820 firmware for the X25-M offers a significant improvement over the old 8160 version, as it manages to maintain performance at a higher level than was possible before. Both throughput and I/O performance are more predictable and closer to the maximum results. Compared to the performance increases caused by driver or firmware updates on graphics cards or for processors and platforms, the improvements here are much more significant. Intel X25-M Mainstream Solid State Drive -... (4 offers) Online shop Price Buy.com $346.82 ElectronicsPLUS $404.99 SuperMediaStore.com $479.99 NextWarehouse.com $610.87 See more products

  • Intel’s latest X25-M drive is based on the same in-house controller with 32MB cache memory as the initial X25-M series, which is why you can keep grabbing firmware upgrades at the usual location. There have been several firmware updates that help prevent substantial performance drops after intensive use. The latest firmware, which we used for the review, does not deliver much higher performance than previous X25-M drives, but the performance impact under intensive use seems to have abated to the point of being almost unnoticeable. Even after significant use of the drive, we were still seeing 200 MB/s read throughput. This latest Intel model carries a “G2” code in its model name, standing for 34nm MLC flash memory, while “G1” marks the older 50nm generation. Both perform equally, but the 34nm version is the basis for the upcoming 320GB model. We found that it required quite a bit of power to deliver workstation-type I/O (2.2W), which spoiled the efficiency results in the I/O performance per watt summary. Almost the same applies to power at maximum streaming (up to 1.8W). However, these results indicate maximum performance. Stay under these levels and you’ll find a 0.2W power requirement for HD video playback as well as 0.1W idle power draw. Only Corsair’s P256 and the OCZ Summit are close at 0.2W; all other SSDs require 0.4W to 1.0W at idle. One watt does make a difference in battery runtime over the long run. However, we’re missing progress on the performance side, as the X25-M is no longer the best flash SSD for performance users. The exception is in enterprise scenarios, which Intel dominates thanks to incredible I/O results. Almost all Indilinx-powered SSDs now deliver higher throughput than Intel. Still, the difference is small once the SSDs are in the 200 MB/s range and up.

  • Performance Zooms with New Firmware for Intel 34nm X25-M SSDs

    10/28/2009: Woops! Intel has pulled the firmware. Check out our news story to find out why. Intel today released new firmware for its 34nm X25-M SSDs. New Firmware: 02HA; Old Firmware: 02G9 The new firmware supports TRIM. This feature requires Windows 7 with Microsoft AHCI drivers. Intel drivers don’t support TRIM yet. For Windows Vista or XP you have to use the Intel SSD Toolbox, which supports TRIM. The TRIM attribute of the ATA data set management command synchronizes the operating system’s view of deleted files with those that are deleted, but not erased on the drive.  TRIM tells the SSD which data blocks are no longer in use. This helps stabilize the performance and health of the SSD over time. For the tests we set up a Windows 7 system on our standard storage test platform. We used the Microsoft AHCI driver to support TRIM. Drive performance is much better with the new firmware. Also after torturing the drive with IOMeter. The sequential write performance in used state areas is in the close to that of the former new state areas, the numbers still decrease but not near as much. Random reads and writes in our database benchmarks double. Very impressive. Workstation benchmarks also double and Webserver and Fileserver numbers increase significantly. Early adopters who bought the first X25-M will not get a firmware update. Very sad. Testing Order: H2benchWIOMeter30min idle time for triming the driveSecond run of H2benchWSecond run of IOMeter The new firmware provides very impressive numbers and takes SSDs to a higher level of performance well beyond HDDs as system drives in desktop computers.