SSD Write Endurance 25nm Vs 34nm

[Christopher]

I believe we are trying most of the suggestions made on your forum, I found "Voodoo science" to be a funny expression that's all.

I'm pretty sure that my Asus M4E-Z (Z68) has the 10.6 OROM, there has been almost no updates on that particular MB and so I'm looking for 11.X as well.
The M4E-Z is the next one one my list if the current setup fails I'll try disabling Hot Plugging, that is probably the last thing I can try on the current OROM that can make a difference.

Maybe I misinterpreted Tony's remarks, but I'm pretty sure he just thought we were just standing around blaming the drive and not looking elsewhere for answers, which is certainly not the case.

I don't think disabling hotplugging matters if you're using the RST drivers... hot plugging is still enabled with RST even if it's turned off in the UEFI. I tried the msahci drivers with Tony's magic voodoo internal-only port hack, the end result of which was having to full power cycle the system (instead of unplugging the drive). But that gave me another idea that I'm working on at the moment. If it doesn't work out, I'm switching back to the Intel board and upgrading the UEFI to the no-longer-available 2053 revision. The readme is no longer available, but the only change was adding more cpu's and updates to RST.

Regardless of motherboard, I do not believe I'd be having these issues if I was only writing 5-10GB a day to the drive.

I was looking forward to getting this drive just to see if the SF BSoD issues would pop up in addition to the LTT question. I'm thinking about getting another toggle nand SF to see what happens if you use it like a normal person :]

[Christopher]

Ao1,

What system did you run your V3 on for 168 hours? Was that consecutive running?

The problems I've been having make me want to try out another SF2x drive, but the problems I've been having make me scared to buy another SF2x drive.

If that makes any sense...

[Ao1]

I've ran the V3 mainly on an ASUS ICH9 mobo, however I also ran it on two laptops and an AMD board. The 168 hours was consecutive and the drive has been hammered quite a bit since then. We should exchange drives to see if I have problems with your SSD and vice versa.

[Christopher]

My drive worked well in my Core2Duo lappy, but the SATA II speeds on it are god-awful. With 46% compression I'd get 96MB/s and if I disabled trim it would be more like 64MB/s after a couple loops. It was stable all the time I ran it, but I rapidly grew impatient.

Actually, if anyone who is in a position to know (???) can tell me if there is a particular model of SF2281 drive floating around out there that is particularly troublesome I'd consider getting one to test. It would seem like all drives would have an equal chance of being problematic (depending on your perspective), but it's possible some configurations with certain FW are worse than others. If someone who has access to that kind of information can point one out to me I'll
pick one up.

[Anvil]

The Force 3 is still running, currently at ~17 hours and still at ~110MiB/s on avg.

Not sure what drive to recommend that is guaranteed troublesome, looks like it's just about being lucky or not.
I've had issues with 60GB and 120GB, none that I know off on the 240GB, it might be due to them being used mostly in raid.

edit:
Both Z68's MB's are on OROM 10.6.0.1091 (ASRock Extreme4 and ASUS M4E-Z)

rst_orom.PNG

[Christopher]

The Mushkin just disconnected when I tried to open RST storage utilities... sheesh.

[Anvil]

That's just typical, it might have triggered something that tipped the scale.

[bulanula]

Innuendo blaming everyone (including end users) and the dog is not helpful and just comes over as an attempt to deflect the problem away from SF2xxx products.



More innuendo with no substantiaition ...oh let me guess NDA

I have updated the firmware on my V3 whenever f/w versions have been released. I ran the endurance app for 7 days straight (~168 hours). I have installed Win 7 and Win 8 in my main PC without any tweaks whatsoever. I’ve installed the V3 as an OS drive in two different laptops, again with zero tweaks. I have used sleep in all instances.

Not a single problem to report. SF2xxx drives seem to either work or not work regardless of the configuration.

Despite the fact that some drives appear to work OK the potential for data loss, combined with the fact that neither SF nor any of the vendors know what the problem is makes the situation unacceptable. Forget trying voodoo, in my view SF2xx products should be recalled.

SF recall will never happen. They are too greedy. Somebody should contact consumer protection LOL

[B.A.T]

Todays update. Another milestone passed
m4
604.5619 TiB
2213 hours
Avg speed 91.12 MiB/s.
AD gone from 09 to 05.
P/E 10544.
MD5 OK.
Still no reallocated sectors


Kingston V+100
117.9872 TiB
464 hours
Avg speed 76.22 MiB/s.
AD gone from 83 to 80.
P/E ?.
MD5 OK.

[Anvil]

Superb progress there B.A.T
> 600TiB is indeed a milestone.

--

Been playing with one of my X58-UD7's and I've upgraded the OROM.

orom_11.PNG

Not sure if there is a RST 11.x yet?

If it's stable I''ll move the Force 3 to this computer If/When the Force 3 disconnects again.
Looks like I'll be running a Force GT as the boot drive.

[B.A.T]

It has endured 92 day with continuous writing and there is no signs of failure. Lets hope it gets past the 700, 800, 900 and 1000 TiB mark

[Christopher]

I think the previous two crashes were actually along the same lines, but with me opening CrystalDiskInfo instead -- wouldn't it be funny if the key to stability was turning off SMART in the UEFI?

On a side note, the fact that the drive is not well behaved is not frustrating on its own... it's frustrating to think that's there might not be anything I can do to correct it.

Anvil, I thought the orom was updated with the BIOS/UEFI... how did you update yours? There isn't an 11.x version of RST yet.

EDIT
I've been wondering about getting another SF2281 to use as a boot drive, or if that would be asking for trouble...


>>Congratulations, B.A.T. <<
I guessed that the M4 would hit 750TB, but it looks like even that was too conservative.

Well Done!

[Anvil]

It is updated with the BIOS/UEFI, it is possible to replace/update ROM's in the BIOS/UEFI using special tools and that's how this is done.

Tony mentioned OROM 11 and so a google was all it took
There are a lot of modders for GB MB's and one can find these modded bioses on the standard sites.
I traced one of the updates to another forum and most GB boards are already modded with OROM 11.

There are no 6Gb/s ports on this MB though, except for a RAID controller but I'm not going to use that one, yet.
It's almost ready so if it disconnects tonight I'll move it to this rig.

[Christopher]

The Mushkin just crashed again, so here's tonight's update... I'm going to have to take the drive down for a while.

Mushkin Chronos Deluxe 60 Update

05 2
Retired Block Count

B1 13 (Down from15)
Wear Range Delta

F1 132555
Host Writes

E9 102199
NAND Writes

E6 100
Life Curve

E7 47 Down from 55)
Life Left

Average 126.55MB/s Avg
RST drivers, Biostar TH67+

333 Hours Work (24hrs since the last update)
Time 13 days 21 hours

11GiB Minimum Free Space 11500 files per loop, 12.9 loops per hour

SSDlife expects 12 days to 0 MWI

[Ao1]

LOL, it took ~ 18 powers on hours for a 1% increment in the Timed Workload Media Wear to appear (E2). In that time I incurred 10.47 GiB of host writes, which means, based on the workload over the last 18 hours, I will put 0.48% wear on the drive per year if I used the SSD 24/7/365.

The Intel Toolbox SMART attributes are showing; 2.01 TB of Host writes after 3,060 power on hours and a MWI of 100%, which equates to 16.14 GiB per 24 hours, so perhaps my use in the last 18 hours was less that average, however anyway you look at it the SSD will outlast me by a long shot.

[DooRules]

there are these alpha version...
http://www.pctreiber.net/2011/intel-...a-version.html

[Tony]

Innuendo blaming everyone (including end users) and the dog is not helpful and just comes over as an attempt to deflect the problem away from SF2xxx products.



More innuendo with no substantiaition ...oh let me guess NDA

I have updated the firmware on my V3 whenever f/w versions have been released. I ran the endurance app for 7 days straight (~168 hours). I have installed Win 7 and Win 8 in my main PC without any tweaks whatsoever. I’ve installed the V3 as an OS drive in two different laptops, again with zero tweaks. I have used sleep in all instances.

Not a single problem to report. SF2xxx drives seem to either work or not work regardless of the configuration.

Despite the fact that some drives appear to work OK the potential for data loss, combined with the fact that neither SF nor any of the vendors know what the problem is makes the situation unacceptable. Forget trying voodoo, in my view SF2xx products should be recalled.

And you wonder why I rarely post on here any more think who I am dealing with, think what they would do if i disclosed everything I know...sometimes you guys live in a strange world where you forget everything is covered by embargo and NDA.

You guys do a good job, you would do better diverting some attention to testing Oroms for stability...there must be bios modders here who can inject the files...see what you find, I have end users doing this now with marked differences in stability.

Don't think the host controller/motherboards always work correctly...they never do, check the release notes with the oroms and RST..check the errata for the chipsets and make sure you are not fighting known bugs.

All I want now is for all motherboard manufacturers to get on the newest and most stable Orom, then I can get FW build that suits and works well with that Orom...and we all move forward. As it is now, im fighting boards on 10.1, 10.3, 10.5 some on 10.6 and a few which generally give no issues on 11.

Regards some of the things you guys have found....we watch what you are doing and like i said you do good work...just don't feel I have to break NDA to justify what I say..I do not and will not.

my view SF2xx products should be recalled.

never going to happen, you guys think the bsod issue is MASSIVE...its not, we have been selling 100k+ drives a month and no where near 100K of them show this issue or come back on RMA. My POLL thread has 160 people tops posting and near 50% of them are having a more positive experience...There is an issue for some and we are working on it..and yes making progress.

I run an MSI 890FXA GD80 here with a 1090T at 3.8, ram at 1200 6-5-5-18 (8GB 1T) memory controller at 2800mhz, main OS drive Revo 480 x2, secondary test OS drives vtx3, vtx3maxIO, Agy3, solid3, Revo3 x2, Revo3...you name it I have run it...and for day to day I have no tweaks, the machine hybrid sleeps (even with the old revo's) and with the single drives I test Hibernation also..and I have never seen an issue except 2 drives disconnecting which I cured with a reg hack to force the sata ports to Internal mode only...so just like Ao1 I run for hours and hours and never see issues....i have not seen what the end users call a BSOD issue on any of the drives I have here and I have 20+ sata3 drives to hand. I tested these drives on Asus p67, Asrok Z68, MSI Z68 etc etc....never seen an issue.

It baffles me, i hoped it was just the drive as then I knew we could fix it...now im thinking its connected to something else that I have no control over.

So here is the challenge guys, join in the search, test the crap out of all the Oroms, all the drivers, all the boards, all the CPU's and all the drives you can...see if you can come up with what causes the boards to disconnect the drives, we can share notes Just don't ask me for what I know to date...as you guessed it...I can not tell

[Anvil]

Kingston SSDNow 40GB (X25-V)

371.45TB Host writes
Reallocated sectors : 10
MD5 OK

35.13MiB/s on avg (~24 hours)

--

Corsair Force 3 120GB

01 92/50 (Raw read error rate)
05 2 (Retired Block count)
B1 54 (Wear range delta)
E6 100 (Life curve status)
E7 72 (SSD Life left)
E9 114262 (Raw writes)
F1 152229 (Host writes)

110.03MiB/s on avg (~24 hours).

power on hours : 439

[Ao1]

My POLL thread has 160 people tops posting and near 50% of them are having a more positive experience...There is an issue for some and we are working on it..and yes making progress.

Yeah, but a number of people posting in the poll to say the problem had gone away where people that had previously stated they had not had a problem and/ or had voted on a previous poll to say the preceding firmware fix had resolved their problem.

There are always going to be bugs out there, it’s a question of the extent of bugs that seem to only affect SF2xxx SSD’s. Plenty of people have reported changing to non SF brands only to see the problem disappear.

Anyway this is going off topic, so I wish you the best of luck finding the problem.

[Christopher]

Tony,

Coming in to this, while I was waiting on my drive to arrive in the mail, I posted a comment about "seeing what this SF2xxx issue was all about." I don't think you learn from things that go right, but rather from the things that go wrong. I don't think the problem is widespread... I don't see craven mobs with torches and pitchforks at banging on manufacturer's doors. Rather, I think some drives just don't work with some motherboards (in this kind of scenario). I think if I was using my SF2281 as God intended, I wouldn't have run into these issues [to the extent that I have]. Truth is, I've learned a lot in just the two weeks I've been testing, and at the bare minimum it's a way to stave off the bordom of unemployment. I'm willing to try anything twice, and while some of what I have read strikes me as "unorthodox", I've kept an open mind about what others (such as yourself) have introduced into the discussion. By all means, if you have any advice, you know where we're at, and I personally would welcome any such guidance, no matter how seemingly ridiculous or sound the advice might appear. But what the hell would I know? Probably not much, but that's why I'm doing this. As I wrote earlier, it's not aggravating that I'm having these problems, it's aggravating that I might not be able to fix it -- but then that's half the fun (and anguish). If I wanted easy, I'd scrapbook or collect stamps.

If you're actually saying that you're having difficulty tracking down cases with these drives, then you can at least add a couple to the mix, but I certainly don't think we're just polishing the brass on the Titanic.


Regards,
Christopher

[johnw]

So here is the challenge guys, join in the search, test the crap out of all the Oroms, all the drivers, all the boards, all the CPU's and all the drives you can...see if you can come up with what causes the boards to disconnect the drives, we can share notes Just don't ask me for what I know to date...as you guessed it...I can not tell

Whether true or not that you signed something saying you could not share information with customers that might help customers to find out FOR YOU why your company's products are defective, such claims are certainly a good reason for people not to buy OCZ products.

[Tony]

Whether true or not that you signed something saying you could not share information with customers that might help customers to find out FOR YOU why your company's products are defective, such claims are certainly a good reason for people not to buy OCZ products.

John...we wait on X79, im hopefull "2281 will just work better all of a sudden"

then we will start asking why it does not work as well on P67 and Z68

if off now, just could not leave the voodoo quote without saying a few words LOL

[johnw]

John...we wait on X79, im hopefull "2281 will just work better all of a sudden"

then we will start asking why it does not work as well on P67 and Z68

So, customers should wait for new chipsets to come out to find out why OCZ's products are defective. Another good reason to avoid OCZ products.

Care to go for a third?

[Christopher]

there are these alpha version...
http://www.pctreiber.net/2011/intel-...a-version.html

So do these only work with gigabyte and asrock?

Scratch that. I've got it working...

Seems to be a good deal faster back on this Intel board with RST 11 Alpha.

[Tony]

So, customers should wait for new chipsets to come out to find out why OCZ's products are defective. Another good reason to avoid OCZ products.

Care to go for a third?

im sorry m8 but i will never get my point over to you...lets agree to disagree and move on ehh?