SSD Write Endurance 25nm Vs 34nm

[Anvil]

The thing is that some of the firmware-enhancements may have altered and bettered both WA and WL.
Some of the drives have had the fw updated multiple times during the test, that leaves some uncertainty about how the drive could have performed if it was using the later/best fw since day 1.
Well, I'm going to update to the latest fw when/if it shows up, as would I in real life.

Anyways, it is interesting seeing what the figures look like as they should be close.

[Christopher]

The 01FW for the 830 is a sack full of pain where WA is concerned. I think the V4s WA got markedly worse with 1.4FW. Most drives don't get that much of a bump one way or the other, and it's hard to know what's going to happen before you flash.

I know that the M3P's GC got tweaked on 1.03 FW, but I didn't see much difference. I did a lot of IOmeter and power consumption testing for several hours before puttin the M3P to work.

[[XC] Oj101]

1.5 PB?!! That Samsung's doing great

Congrats on becoming a mod Anvil, when did that happen?

[canthearu]

The Torqx 2 had its first bad block at approximately 40600 GiB, with 1677 average NAND cycles, 2539 max NAND cycles.

The Vertex 4 had its first bad block at 294.27 TiB Host writes.

If you have an octane/petrol or vertex 4 and your drive gets a bad block, you better have a backup!

[Anvil]

The Kingston (X25-V) had reallocation's (1) from the get-go.

I'm sending a list of all reallocation's to Ao1. (until SmartLog was introduced)

[Christopher]

I've actually moved the two drives out of the endurance rig and into my main desktop. I want to see what happens over the next week, as I noticed that the drives seem to be not in-considerably faster.

After 21 hours, the M3P is at 270MB/s and the 830 is at 303MB/s. I'm probably hallucinating, but I feel like there is something preventing the drives from running quite as fast as they should. Both rigs are on 10.6.1002, one H67 and Z68.

Perhaps I imagined it, but we'll see.

[canthearu]

Both rigs are on 10.6.1002, one H67 and Z68.

I am using 11.1.0.1006 on my Z77. Seems to be working pretty well.

Could power management be a factor.

Maybe you are wearing out the SATA cables with all the transferred data

[Christopher]

I am using 11.1.0.1006 on my Z77. Seems to be working pretty well.

Could power management be a factor.

Maybe you are wearing out the SATA cables with all the transferred data

Maybe all those PB of host traffic are wearing out the PCH... but no, I don't think so. Not sure what it is, but when I was doing preliminary testing I would use my desktop, and it would be around 270 avg for the M3P. I'd put it in the e-rig, and it would do 240ish MB/s. Both seem to be doing much better, but it will be a few days before I know for certain.

[Christopher]

Plextor M3P 128GB Day 1

27,386.66 GiB
26.74 TiB

269.65 MB/s Average

25 Hours

Rellocated Event Count 0
--------------------------------------------

Samsung 830 256GB Day 70

1,683,657.62 GiB
1,644.20 TiB

304.48 MB/s Average
1667 Hours

7260 Wear Leveling Count
MWI 1

6/0 Erase/Program Fail
Used Reserved Block Count: 12/24572 sectors

--------------------------------------------

The M3P is doing 269MB/s avg after 25 hours, the 830 is doing a staggering 304MB/s. We'll see where they settle.

[Anvil]

Are the rigs set up differenty?
(drivers, AV scanning, ...)

The speed of the M3P almost looks too good to be true, how many days will it last based on that speed. (based on i.e 3K P/E cycles)

[Christopher]

Are the rigs set up differenty?
(drivers, AV scanning, ...)

The speed of the M3P almost looks too good to be true, how many days will it last based on that speed. (based on i.e 3K P/E cycles)

The rigs are very similar, neither has AV or anything like that -- especially the endurance rig, which is about as stripped down as a Win7x64 install can get. The instantaneous speeds are higher, not just the average.

The M3P 128GB is a fantastic drive. Having 32 dice helps. It maxes out at 350MB/s writes at just about every transfer size.





I think the 128GB 830 would be almost as fast, but the 830 is a little more inconsistent. The M3P does 322MB/s through the entire loop just about.

The two drives could be fast enough that other factors are coming into play with regards to speed. At 269.73MB/s avg, you'll get 22,758GiB a day.Each day the drive will use about 197 PE cycles based on 1.1 WA

[Anvil]

Intel 330 120GB

46.92TB Host writes
Reallocated sectors : 05 0
Available Reserved Space : E8 100
MWI 96
[F1] Total LBAs Written 1537505
[F2] Total LBAs Read 22623 // no movement
[F9] Total NAND Writes 33879GB
POH 144
MD5 OK

122.66MiB/s on avg (~48 hours)

--

Kingston SSDNow 40GB (X25-V)

990.32TB Host writes
Reallocated sectors : 05 46
Available Reserved Space : E8 99
POH 8869
MD5 OK

33.99MiB/s on avg (~48 hours)

[B.A.T]

Todays update:

m4
957.0306 TiB
4027 hours
Avg speed 74.35 MiB/s.
AD gone from 66 to 63.
P/E 16483.
MD5 OK.
Reallocated sectors : 00


Intel X25-M G1 80GB
658.64 TiB
23339 hours
Reallocated sectors : 162
Available Reserved space: 90
MWI=124
MD5 =OK
42.82MiB/s on avg


Intel X25-E 64GB
138.93 TiB
426-30=396 hours
Reallocated sectors : 0
Available Reserved space: 100
MWI=99
MD5 =OK
101.07 MiB/s on avg

[Christopher]

Avg speed on the 830 is down 2%, M3P is up by 1%.

After 37 hours, it's just about blown through 10% of the PE cycles! Exciting.

I wish the MWI implementation on the Plextor wasn't so damn lazy.

[Anvil]

I haven't gotten to test mine yet, I've updated the firmware so they are ready.

Have you got a recent screenshot showing the M3Ps SMART status/values?

[Ao1]

Here are a couple of charts showing relocated sector events for the intrepid X25-V. I had forgotten that relocated sectors turned up on the X25-V before the MWI had expired. It follows the bath tub curve quite well. (Infant mortalities, stability and then progressive failure).

[Christopher]

I haven't gotten to test mine yet, I've updated the firmware so they are ready.

Have you got a recent screenshot showing the M3Ps SMART status/values?

Certainly. I'll post it later, but all current/worst values are nominal. Only power on/cycles and host writes are changing.

I'll post it when I get home.

[kaktus1907]

I'm really hoping it will appear to be 5K P/E(Toshiba Type B 24 nm T-Nands).. sadly corsair performance pro's smart values arent any different than plextors

[Christopher]

I'm not sure how upset I would be if I bought a Performance Pro and I got 24nm instead of 32nm. While it is uncouth to swap flash without saying so, the 24nn uses much, much less power. The M3P maxes out at 2.6w, and it doesn't even get above ambient during endurance testing. If it puts in a halfway decent showing, the 24nm might be what you want.

We will see how the M3P does, but previous drives using Toshiba toggle have not done well at all (Samsung's Toggle appears to be a different story).

[Anvil]

Thanks Ao1,
Hopefully they'll be able to make "drives" like that again.
It has been uneventful except for enduring more than any other drive, so far.

Christopher,
Just wait until something interesting happens.

[Christopher]

M3P Smart


So power cycles, power on hours, unsafe shutdown and host writes are the only values which increment so far.

I expect if/when a block goes bad MWI and reallocations will change.

All current/worst values are still at 100.

[Anvil]

474556*64MiB that's close to 29TiB

I expect CDI and or others will pick up on the writes attribute, lets hope reallocations won't happen for a while, at the current speed it probably won't take more than a few weeks though.

[Christopher]

474556*64MiB that's close to 29TiB

I expect CDI and or others will pick up on the writes attribute, lets hope reallocations won't happen for a while, at the current speed it probably won't take more than a few weeks though.

That was a shot from last night. It's currently at 714341 (44,646 GiB). I just didn't want anyone thinking that the attribute didn't work correctly (it does -- B1 RAW * 64 = host writes MiB).

[canthearu]

That M3P is in a real hurry. I'm not sure where it is in a hurry to get to, but where-ever it is going, it won't take very long.

[Christopher]

That M3P is in a real hurry. I'm not sure where it is in a hurry to get to, but where-ever it is going, it won't take very long.

It's going to Hell, and quickly at that. I'd expect to maybe see a reallocation in around two weeks -- somewhere near 3000PE cycles or so. I hope I'm wrong though. Fortunately for me, I usually am.

Maybe Plextor's price premium is in part due to higher binned NAND, perhaps analogous to the Kingston HyperX 5K? Doubtful, but I remain hopeful. I think they just charge more because it's fast.

--------------------------------------------

Plextor M3P 128GB Day 2

49,245.08 GiB
48.09 TiB

271.21 MB/s Average

48 Hours

Rellocated Event Count 0



--------------------------------------------

Samsung 830 256GB Day 72

1,707,352.20 GiB
1,667.33 TiB

300.36 MB/s Average
1694 Hours

7260 Wear Leveling Count
MWI 1

6/0 Erase/Program Fail
Used Reserved Block Count: 12/24572 sectors


--------------------------------------------