SSD roundup: Vertex 3 vs M4 vs C300 vs 510 vs 320 vs x25-M vs F120 vs Falcon II

[koc]

Koc, pls keep us updated cause i plan to get the same drive (but 128gb).

sorry for delay ,
I upgraded to 0002 firmware and the problem gone , no more freeze issue , well done crucial

[Pillo-kun]

About a day ago i bought a Corsair Force GT 120GB and had several freezings that resulted in BSOD's. thought maybe that the mems was the cause of it but it turned out to be the ssd... gonna return this evil sucker and wait a while longer for a ssd that works. Ihope that the store dont cause hell because i already returned a psu that had coil whine when the pc was off.

[bluestang]

Here is 2x 64GB M4 in RAID0, 128k Stripe on P67 PCH, i7-2600k running all cores at 3800MHz

[Mech0z]

I am getting a m4 128gb tomorrow (Replacing my old vertex 1 60gb) how do I check if ti has the newest firmware?

[icecpu]

I am getting a m4 128gb tomorrow (Replacing my old vertex 1 60gb) how do I check if ti has the newest firmware?

On the label it will say firmware 001 or 002 is the latest

[Mech0z]

Super I hope I dont have any issues with it in my Thinkpad W520 should be a bit better than my current Vertex 60Gb think that drive is nearing its death, I have some wierd stops in windows

[Computurd]

if i were to buy today i would pick up an M4, some of that has to do with the Xtreme reliability that i have had with the 8 c300s and their sustainable performance. They are just solid as a rock, no variability when it comes to different compression factors.
just rock solid performance and reliability. what more could one want?

[RealTelstar]

So is the Crucial M4 still the best choice? No known issues?

Yes i think so, with firmware 002 i heard no issues from anyone.

[DooRules]

seems like vast majority of problems with M4 is with laptops, and with those Macs

[trans am]

bluestang those numbers are awesome. I just picked up 2 of the same drives and the 64gb c300 I already had and running them on the lsi 9260-4i with fastpath key and I am not getting those reads. I have the lsi plugged in on the last x16 slot so that runs at 4x so that might explain mine. Or its that the c300 isnt playing nicely with the m4's. I need to get another m4 asap.

[Chewbenator]

Does it make sense to buy an M4 now or wait for new hardware?

I'm building a computer sometime in the next 4 months depending on several release dates and reviews. I could upgrade a few things now, like SSDs that would benefit my current machine. Is there anything coming out in the next quarter worth waiting for in terms of SSDs?

[trans am]

Does it make sense to buy an M4 now or wait for new hardware?

I'm building a computer sometime in the next 4 months depending on several release dates and reviews. I could upgrade a few things now, like SSDs that would benefit my current machine. Is there anything coming out in the next quarter worth waiting for in terms of SSDs?

I would get a pair of 64gb m4's now. They are around $110 ea. keep the boxes and sell them when the time comes or when you get the rig just keep adding on as the prices get lower and lower. If you never had ssd or ssd in raid0 then prepare yourself for the most noticeable upgrade you've ever seen! Screw the new rig in 4 months you could be dead. carpe diem! cease the day! you will be glad and 64gb drives are easy to just keep adding on and not too expensive.

[Chewbenator]

Saw a great deal for 128GB M4s for $160 + tax, so picked a couple up. Thanks for the helpful thread guys.

Here's the deal if you are wondering: http://www.buy.com/prod/crucial-ct12...221150373.html

[deathman20]

Saw a great deal for 128GB M4s for $160 + tax, so picked a couple up. Thanks for the helpful thread guys.

Here's the deal if you are wondering: http://www.buy.com/prod/crucial-ct12...221150373.html

Dang that is a heck of a price for that drive. You say your building a new rig in a few months but want the drives now? Get the drives and use them til the new rig comes? Or just buy them at that time since prices will be either dropping or something new might be out... doubtful but could happen. If anything we might start hearing about the next drives in the line up by then.

[Chewbenator]

I bought the drives because I know I will notice the performance increase immediately, and I don't have to upgrade anything else. I'm also replacing the CPU, MOBO, PSU, GPU(s), Monitor, and RAM. Other than the ram everything else would require something else to be upgraded as well.

Back on topic, going to use onboard raid-0 and judge the performance from there.

[RealTelstar]

Back on topic, intel's new firmware does NOT fix the 8gb bug of the 320.
Therefore, the crucial m4 and intel 510 are the only reliable drives currently available.

[therat]

Do either the M4 or Intel 510 suffer from the "sleep/hibernation" bug? This is important to me and the main reason I want to upgrade from my Vertex 2.

[Wendy]

Back on topic, intel's new firmware does NOT fix the 8gb bug of the 320.
Therefore, the crucial m4 and intel 510 are the only reliable drives currently available.

Can you tell us all a little bit more about the new 320 firmware NOT fixing the 8MB bug?

Also, I think something is amiss with P55/P67/Z68/H67, as the Intel 510 is not that stable either, and since the M4 shares the same controller, I would imagine it is also affected.

We have SandForce SF2281, Marvell, and now Intel 320 all having problems, and all seem to suffer an "unexpected power loss".

Here is what I found with Intel 510 and OCZ Vertex 3.
Read the "stability section"
http://www.myce.com/review/intel-510...conclusion-11/

[Ao1]

Wendy I have not had time to follow (in detail) the issues being reported on various SSD's, so when I ask these questions I am not trying to have a dig; rather I am trying to get myself up-to-date. First off, why are Intel stating they don't plan to develop a fix for the 15.Intel ME Clock Throttling Failure Causes Hang bug? Does the hang only occur during post, or does it cause hangs after the system has posted? I'm struggling to understand why Intel have taken the position to not bother developing a fix. How does this issue relate to the problems being reported by SF2xx drives? Does it impact other storage systems?

Regarding the 27 page thread on "Intel Management Engine cmos corruption and its effects" on the OCZ forum can you please confirm the outcome? Has corruption been confirmed? Is it reproducible? In what way does it impact storage systems? Does it only impact SF drives or does it impact any storage device?

How are the BSOD/ freezing events being reported on SF2 drives able to manifest themselves over a wide range of platforms including SATA 2, SATA 3, PCIe, AMD chipsets, Intel Chipsets and dedicated controllers? Why do most (all?) people report that BSOD/ freezing problems disappear when they revert to a HDD or a non SF drive?

The 320 “BAD_CTX 0000013x” issue is caused by an unexpected power loss under specific conditions. What those specific conditions are I don't know. I believe it was first discovered by someone repeatedly removing and then replacing the power supply. I don't think it is in any way related to the SF problems but without knowing the specific conditions it is hard to say. Since the f/w update I haven't seen any reported reoccurrences, although I haven't looked that hard either.

In your review I'm not sure if you are saying that the 510 had issues or if the motherboard had issues. Is it one or the other or both?

On a final note, with regards to your test procedures, did you precondition the drives or where the drives in a fresh state when you benchmarked them? Why did you choose to compare a 240 GB drive against a 120 GB drive?

[Wendy]

There is a lot of questions in there Ao1, and I don't have definate answers to most of them. That is why I said "I think" P55/P67/Z68/H67 has a problem.
Let's have a go anyway.

Wendy I have not had time to follow (in detail) the issues being reported on various SSD's, so when I ask these questions I am not trying to have a dig; rather I am trying to get myself up-to-date. First off, why are Intel stating they don't plan to develop a fix for the 15.Intel ME Clock Throttling Failure Causes Hang bug? Does the hang only occur during post, or does it cause hangs after the system has posted? I'm struggling to understand why Intel have taken the position to not bother developing a fix. How does this issue relate to the problems being reported by SF2xx drives? Does it impact other storage systems?

IMO, there are at least two ways of answering this question.
First, this a platform problem, and is related to the system as a whole.

1, Intel don't think it's important enough to warrant a fix.
I find this highly unlikely, as IME forms the hub of the chipset, among other things it controls power management, and link management around the board.

2, There is an error in the hardware, and these chipsets are to far into production to fix without another recall.
Note Intel state that the may try and work "around" the problem with an IME firmware update.

It may hang at POST, but IME also needs to be working properly during a user session. If IME isn't working then advanced power states also wont work properly, and link power states wont work properly.

Regarding the 27 page thread on "Intel Management Engine cmos corruption and its effects" on the OCZ forum can you please confirm the outcome? Has corruption been confirmed? Is it reproducible? In what way does it impact storage systems? Does it only impact SF drives or does it impact any storage device?

The IME table corruption was confirmed privately, then an NDA applied.
I could reproduce it at will with UEFI 1.4, it was fixed in UEFI 1.5, and remains so with UEFI 1.6.

How are the BSOD/ freezing events being reported on SF2 drives able to manifest themselves over a wide range of platforms including SATA 2, SATA 3, PCIe, AMD chipsets, Intel Chipsets and dedicated controllers? Why do most (all?) people report that BSOD/ freezing problems disappear when they revert to a HDD or a non SF drive?

BSOD is not uncommon on P67/Z68 with just an HDD attached. One of the problems (as I see it) As soon as a person adds an SF2281 based SSD and they have a BSOD, they automatically blame the SSD for the BSOD. This is not always the case, and this has been proved by many people who had BSOD problems with an SF2281 being able to get their systems completely stable, once the real cause of their BSOD had been fixed.

The 320 “BAD_CTX 0000013x” issue is caused by an unexpected power loss under specific conditions. What those specific conditions are I don't know. I believe it was first discovered by someone repeatedly removing and then replacing the power supply. I don't think it is in any way related to the SF problems but without knowing the specific conditions it is hard to say. Since the f/w update I haven't seen any reported reoccurrences, although I haven't looked that hard either.

I've read many people who had the 8MB bug appear just switched their PC off, and when they powered their system up the next time the bug appeared. They hadn't suffered a power loss to the system.

What I do know. The Intel 510, Vertex 3, and C300 all show the same error event in S.M.A.R.T. and that is an unexpected power loss to the SSD, and I can say for certain, at least on my system, the PC never suffered a power outage of any sort.

In your review I'm not sure if you are saying that the 510 had issues or if the motherboard had issues. Is it one or the other or both?

Neither. I feel the problem is either with the chipset itself, chipset firmware (IME), or perhaps even a driver problem.

Intel has updated the OPROM firmware 3 times already for Z68, and god knows how many RST driver revisions. I doubt Intel are doing this for the fun of it.

I asked Intel for a comment on what I had seen with my 510 review sample. They declined to comment.

On a final note, with regards to your test procedures, did you precondition the drives or where the drives in a fresh state when you benchmarked them? Why did you choose to compare a 240 GB drive against a 120 GB drive?

I used what I had available at the time regarding SATA3 SSDs.

[Ao1]

OK thanks. Just for clarify Intel state that for that particular erratum "no functional failures have been seen due to this issue" - "Status: No Plan To Fix". Are you saying there are functional failures and it will have to be fixed?

I've not heard of anyone having stuttering problems with the 510. You say that the 510 stuttered on 15 different occasions resulting in an unsafe shutdown count being recorded each time by SMART? That is repeatable on all P67, Z68, and H67 boards?

[Wendy]

OK thanks. Just for clarify Intel state that for that particular erratum "no functional failures have been seen due to this issue" - "Status: No Plan To Fix". Are you saying there are functional failures and it will have to be fixed?

To be totally honest, I'm 13 weeks into trying to nail down the problem, and I'm no closer to doing so now than I was 13 weeks ago.
Sandforce and OCZ are laying at least part of the problem that SF2281 faces is down to the chipset, and they are trying to work around those problems. Crucial claimed something similar, saying that LPM is the issue, and worked around that. Intel in the above errata claim that no failures are down to errata 15. It will all eventually come out in the wash.

The reason I was looking for more info on the new 320 firmware is, I suspect this could be another SSD affected by this same issue. Again, this will eventually all come out in the wash.

I'm trying to keep an open mind on things. I suspect their are faults on all sides (P67 etc etc) not behaving as good as it could, and SF2281, Marvell, Intel SSD controllers perhaps not playing well either. It will eventually all come out in the wash.

I've not heard of anyone having stuttering problems with the 510. You say that the 510 stuttered on 15 different occasions resulting in an unsafe shutdown count being recorded each time by SMART? That is repeatable on all P67, Z68, and H67 boards?

Some of these errors were when the 510 was connected as a spare during benchmarking. This didn't cause a stutter as it wasn't the system drive. During later stages of testing, the 510 was the system drive, and then the unexpected power out did cause a stutter.

Not all boards are affected. I have a AsRock Z68 Extreme 4 which does show some problems, a GigaByte P55, and an Asus P67 board that doesn't show the problem. My technical advisor has many boards, and many boards the same model. Out of 9 Z68 boards, he has 4 identical Asus boards, 3 of them are fine with the 4th board showing the problem.

Anand has been able to get the SF2281 BSOD problem to show its face, but can only do this on one of the boards he tested, and now can't replicate the problem on that board. This is the problem with these issues. They tend to be illusive to replicate, and pin down the real problem.

[therat]

so we have a better chance of a stable system if we stick with the X58 chipset boards until the problem is finally resolved?

[RealTelstar]

Can you tell us all a little bit more about the new 320 firmware NOT fixing the 8MB bug?

I just heard users reporting issues after they applied the new firmware.

Also, I think something is amiss with P55/P67/Z68/H67, as the Intel 510 is not that stable either, and since the M4 shares the same controller, I would imagine it is also affected.

I think it depends on the implementation of the controller too. Crucial fixed the stuttering completely with firmware 002. Why intel didnt?
The 8MB bug seems confined to the intel 320, afaik.

Lastly, did you enable hotplug in your mainboard when testing the 510?
I really hope intel fixes the chipsets because it's a real shame. And i dont want to be forced to upgrade to the X79 to have a stable system.

[zalbard]

I think it depends on the implementation of the controller too. Crucial fixed the stuttering completely with firmware 002. Why intel didnt?

What stuttering (that Intel drives have) are we talking about?