SSD Write Endurance 25nm Vs 34nm

[Computurd]

@ct
Wishful thinking?
I bet you won't be booting off that old array unless it's for sentimental reasons

i agree anvil, man i did have some good times with that array...i just cant break them up for some reason.
dammit im getting misty eyed....

[Anvil]

274.34TB Host writes
Reallocated sectors : 6
MD5 OK

34.41MiB/s on avg (23.6 hours)

--

I'm including a screenshot from SSDLife but I guess I need to get some answers on how to configure the app for optimal results/readings.
It looks like it needs to run 24/7 in order to get correct results on "Throughput" for Today and Yesterday, "Last 7 days" looks to be correct though. (it started counting when the app was first started, 1.4TB looks to be correct)

It also reports that there is 39.9GB "free size" which is not correct, so a small bug on one of the drives, strange as the other SSD on the system is correctly reported.
Would have been interesting running this app from day 1.

SSDLife_2011-08-30-11-57.png

[bulanula]

It really is impressive that the M4 reached 400 TB so soon. I wonder how the C300 will fare with same controller and better 34nm NAND. Maybe 600 TB ?

[bluestang]

M225->Vertex Turbo 64GB Update:

176.02 TiB
613.9 hours
MWI 15 (drops by 1 for every 50 raw wear)
4266 Raw Wear
110.53 MB/s avg for the last 15.85 hours (on W7 x64)
MD5 OK
C4-Erase Failure Block Count(Realloc Sectors) at 2 (Bank 6/Block 2406; Bank 3/Block 3925)

[B.A.T]

m4 update:
412.3073 TiB
1350 hours
Avg speed 90.03 MiB/s.
AD gone from 119 to 115.
P/E 7231.
MD5 OK.
Still no reallocated sectors



Kingston update:
The V+100 dropped out again this morning.
It looks like this when that happens.



Last record states:
26.4734 TiB
103 hours
Avg speed 77.37 MiB/s.
AD gone from 175 to ?.
P/E ?
MD5 OK.

[SynbiosVyse]

100% (Incompressible)
102.80 hours
121.8470 TiB written total
40.90 MB/s
MD5 ok

05: 0
B1: 44
E7: 62%
E9: 63040
EA/F1: 125312
F2: 128

[bulanula]

I think the M4 will be the next to follow the Samsung 470 in the "dead" SSD closet.

[devsk]

I think the M4 will be the next to follow the Samsung 470 in the "dead" SSD closet.

It doesn't looks like so from the data.

[B.A.T]

Update:
I'm not able to update my m4 with new firmware. I get this error message:


I've tried IDE with port 1-4, and AHCI on the same ports.

For the time being I'll run with v002 until I find a fix.

The Kingston is also up and running. I changed port on the motherboard and Anvil helped me recontruct the log.

[Vapor]

I'm kinda leaning toward leaving the m4 on 0002...seems late in the game to make a change like that. Of course, not my drive, just my opinion And I don't think it will effect things much at all, either way.

[B.A.T]

I'm kinda leaning toward leaving the m4 on 0002...seems late in the game to make a change like that. Of course, not my drive, just my opinion And I don't think it will effect things much at all, either way.

For the time being I'll stay with the 0002. The update won't work and I haven't found a fix.

[Vapor]

C300 Update

284.34TiB host writes, 4 MWI, 4801 raw wear indicator, 2048/1 reallocations, 63.2MiBsec (going up again...), MD5 OK




SF-1200 nLTT

150.625TiB host writes, 106.125TiB NAND writes, 49 MWI, 1698 raw wear (equiv), wear range delta 3, 55.85MiB/sec (going down), MD5 OK

[devsk]

C300 Update

284.34TiB host writes, 4 MWI, 4801 raw wear indicator, 2048/1 reallocations, 63.2MiBsec (going up again...), MD5 OK

C300 seems to be in bad shape. it has the 34nm NAND. M4 has 25nm NAND, more writes and no reallocations. Something is not adding up.

[Computurd]

M4 has 25nm NAND, more writes and no reallocations. Something is not adding up.

the drive life of the devices with the new nand are supposed to have longer life than previous generations. This is due to more advanced underlying technologies. This is not an unexpected result imo. Optimized controllers is where it is at!

[Vapor]

Intel X25-V had 4 reallocated blocks before the Intel 320 had any....and now the 320 has 69 compared to just 6 from the X25-V. I'm not sure reallocations mean anything until the counter really begins to move consistently. So far, only the Samsung 470 and the Intel 320 have had any sort of consistent increases with the reallocation counter.

The C300 has 1 reallocation (and hasn't moved for nearly 200TiB) and the m4 has 0 reallocations...in addition to their respective initial bad blocks (bad blocks found by Crucial during burn-in). Not much to go off of there, if the other drives are any indication

That said, I do expect the m4 to die a little sooner (or a lot sooner if you go by write days--m4 is noticeably faster). WA is a little higher and the NAND is rated for only 60% of the lifetime the C300's NAND is rated.

m4 has effectively worn out its MWI twice whereas the C300 hasn't even worn it out once. Despite that, I feel both the C300 and the m4 still have many days left in their lifetime....the Samsung 470 effectively wore out its MWI 8 times and it took ~5x MWI lifetimes for the reallocations to begin to move consistently.

[bulanula]

Intel X25-V had 4 reallocated blocks before the Intel 320 had any....and now the 320 has 69 compared to just 6 from the X25-V. I'm not sure reallocations mean anything until the counter really begins to move consistently. So far, only the Samsung 470 and the Intel 320 have had any sort of consistent increases with the reallocation counter.

The C300 has 1 reallocation (and hasn't moved for nearly 200TiB) and the m4 has 0 reallocations...in addition to their respective initial bad blocks (bad blocks found by Crucial during burn-in). Not much to go off of there, if the other drives are any indication

That said, I do expect the m4 to die a little sooner (or a lot sooner if you go by write days--m4 is noticeably faster). WA is a little higher and the NAND is rated for only 60% of the lifetime the C300's NAND is rated.

m4 has effectively worn out its MWI twice whereas the C300 hasn't even worn it out once. Despite that, I feel both the C300 and the m4 still have many days left in their lifetime....the Samsung 470 effectively wore out its MWI 8 times and it took ~5x MWI lifetimes for the reallocations to begin to move consistently.

Well said. I too believe the C300 will outlast the M4 by a considerable amount.

[Anvil]

277.07TB Host writes
Reallocated sectors : 6
MD5 OK

33.27MiB/s on avg (~48 hours)

SSDLife_2011-08-31-12-34.png

[bluestang]

M225->Vertex Turbo 64GB Update:

183.56 TiB
628.8 hours
MWI 12 (drops by 1 for every 50 raw wear)
4400 Raw Wear
130.91 MB/s avg for the last 16.78 hours (on W7 x64)

Not sure why, but this thing was flying last night
That's 20 MB/s more than what it's been doing lately

MD5 OK
C4-Erase Failure Block Count(Realloc Sectors) at 2 (Bank 6/Block 2406; Bank 3/Block 3925)

[B.A.T]

m4 update:
419.5727 TiB
1374 hours
Avg speed 92.81 MiB/s.
AD gone from 115 to 111.
P/E 7356.
MD5 OK.
Still no reallocated sectors


30.8181 TiB
120 hours
Avg speed76.26 MiB/s.
AD gone from 175 to 168.
P/E ?.
MD5 OK.



Changing port on the motherboard seem to fix the problems.

[deathman20]

Wow... That drive is still ticking. So surprised by that 64Gig M4 there.

[B.A.T]

It's a really great drive And with the new firmware it's hard not to recommend it.

[Vapor]

A bit in a rush tonight, so no screenshots and will have to hold off charts for another day.

C300 Update

290.255TiB host writes, 2 MWI, 4900 raw wear indicator, 2048/1 reallocations, 63.25MiB/sec, MD5 OK...next update and MWI will finally be exhausted


SF-1200 nLTT

156.563TiB host writes, 110.813TiB NAND writes, 47 MWI, 1773 raw wear (equiv), wear range delta 3, 55.85MiB/sec, MD5 OK

[SynbiosVyse]

132.69 hours
126.0551 TiB written
40.93 MB/s
MD5 ok

05: 0
B1: 47
E7: 59%
E9: 68224
EA/F1: 129664
F2: 128

[Vapor]

What's MWI at?

[SynbiosVyse]

Sorry I had a long day, I forgot.
59%