SSD Write Endurance 25nm Vs 34nm

[B.A.T]

Hopefully I'll be home and ready to start around 20:00 GMT+1. If that is to early I can wait a couple of hours.

[Christopher]

That should be perfect. This is going to be awesome.

I'm shaking down the test rig right now. I've put the components into a Lian Li PC V351 Black aluminum uATX enclosure.

For the record, this system includeds the following components:

Biostar TH67+ H67 1155 uATX motherboard.
Intel Celeron G530 2.4Ghz dual core 1155 cpu
2x2GB GEiL CAS7 DDR3
OCZ ModXstream Pro 500W semi-modular PSU
Win7 x64 on a Hitatchi Travelstar 40GB 2.5" HDD (on a SATA II port)
Intel RST drivers.
Scythe Shuriken Rev B. CPU HSF

All crammed into the Lian Li V351

[Anvil]

Should be great fun!

I almost envy you but don't worry, I'll find something to test sooner or later...

[minpayne]

I ran Anvil's Storage Utilities for 6 hours on one of my Crucial M4 512GB with 0002 firmware. The wear leveling count was increased from 2 to 8. (Not sure how M4's MWI works)

I'll see if I can find a spare SSD to join this thread

[Christopher]

I ran Anvil's Storage Utilities for 6 hours on one of my Crucial M4 512GB with 0002 firmware. The wear leveling count was increased from 2 to 8. (Not sure how M4's MWI works)

I'll see if I can find a spare SSD to join this thread

Wow, I bet the 512 was haulin' in writes. I'm guessing somewhere near 300MBs.

EDIT
I found a bunch of cheap Patriot refurbs. I'm trying to figure out what is what as they're de labeled and the specs are wrong or non-existant, and they have special refurbished part numbers. I think one is a SF 1200, one is an Indilinx, and the others are a mystery.

But they're cheap, and I don't see why I couldn't throw another SSD in the Chamber of Doom with the Samsung. But no J.Microns are allowed, so I have to do some research. I was thinking maybe the Indilinx was 5xnm Samsung, and I could do it Bluestang style.

[Computurd]

(1) You did not disclose that you are a biased source on the subject of thessdreview, as a responsible journalist would

the people who need to know, do. I didnt think that i was making any statements that would require such disclosure though.

He made a statement, it was right. i merely pointed that fact out. When something is a known, proven fact, not sure that bias matters.

Today is December 4th, 2011.

Now, do i need to disclose that I read that fact off of a calendar that i also wrote on?

i once submitted an opinion article to a newspaper, and they published it. so do i need to disclose that each and every time i discuss that newspaper?

if it will assuage your tender feelings for being PC, then yes, everyone, i have posted evaluations on that site.

I also do consulting and independent analysis for several companies. Did i need to tell you that? I actually make money at those though....

Internet tech journalism has never been very good, but thessdreview stands out as the worst of the worst in this regard.

Your opinion. I also hold that opinion about several sites.

However the site in question is very successful, and growing very fast. there are plenty of other sites for you to go read im sure. and there are plenty of people who enjoy reading thessdreview site.

it is interesting that you tell people to read it with a grain of salt, when the info presented is correct.

(2) My main complaint with the original article, as I stated here previously, was that the author apparently had a confidential source for the information, but rather than stating in the article that "this information comes from a confidential source who we vetted and have reason to believe is genuine" as a responsible journalist would, the author made no statement of the kind but rather listed a couple very flimsy pieces of evidence.

I dont recall you saying such. I think that you said it wasn't true/impossible? shoddy journalism as it was unsubstantiated in your opinion, even though it was clearly speculative. Or that might have been someone else tbh i dont care enough to go look, but i think that was more along the lines....

You are assuming that he did have a confidential source. I surely cannot answer to that. Asking the wrong guy...but i point you to his latest article that brings up this whole issue, where he names his source. so i assume if there were a direct source it would have been named, as he did with the latest intel spec sheets/etc.

if you are in 'circles' though I will say this, lots of information does fly around, none of it official, and alot of it rumors. If you were to hear a rumor from someone in the buisness, and looked at the published specs, yeah it might make sense. simple as that.

Wiki on journalism would nail a situation like that closer to a "lead", but i assure you I have no desire to talk journalism.

A confidential source would be something entirely different *in my opinion*. That would be like someone actually in that organization leaking specs or info.

as long as you control and balance it well!

if only that were so....

[2f_tfe]

very interesting thread, i'm reading from post #1 but i joined a few days ago,
thanks for the great job

luke

[kensiko]

That should be perfect. This is going to be awesome.

I'm shaking down the test rig right now. I've put the components into a Lian Li PC V351 Black aluminum uATX enclosure.

For the record, this system includeds the following components:

Biostar TH67+ H67 1155 uATX motherboard.
Intel Celeron G530 2.4Ghz dual core 1155 cpu
2x2GB GEiL CAS7 DDR3
OCZ ModXstream Pro 500W semi-modular PSU
Win7 x64 on a Hitatchi Travelstar 40GB 2.5" HDD (on a SATA II port)
Intel RST drivers.
Scythe Shuriken Rev B. CPU HSF

All crammed into the Lian Li V351

I hope you don't want to use this computer without a SSD :P

[One_Hertz]

619TB. No other changes. MD5 OK.

[Hopalong X]

619TB. No other changes. MD5 OK.

One day it will throw in the towel.
A resistor or some minor part will give way.
Until then the tortoise continues the race while the hares keep falling by the wayside.

[Christopher]

I hope you don't want to use this computer without a SSD :P

That's a good catch. I'm waiting for the UPS man to bring me my new testing drive. BAT and I are going to be locked in mortal combat in an SSD cage-match from hell. It's all very exciting.


Don't knock the Chamber of Doom. The Celeron G530 is a thoroughly decent CPU when paired with an H67 board. It sips power, has good-enough SB graphics, and has enough zip to do what I need it to do and then some. The single core Celeron might be enough for running the endurance test and remote desktop apps, but the single core doesn't clock down. It actually uses more power overall and costs almost the same amount of money. Actually, the G530 is pretty much good enough for most everything. It's probably almost as fast as a Core i3 5xx 1156 processor.

Also, the black aluminum Lian Li case will probably co-ordinate rather well with the svelt 7mm Samsung 830 chassis...

[Christopher]

I don't know what's going on with the QD4/32 write results. Me like.


I'm starting testing on the drive in about 30 mins, which I believe is 20:00 for B.A.T.

[bluestang]

M225->Vertex Turbo 64GB Update:

Soooo, I started the test back up this morning after letting it sit idle over the weekend...It ran for 104 loops/3.63 hrs at 117.61 MB/s and C5 jumped from 7 to 15 in that time. MD5 in ASU still showed OK. Stopped test as I needed to reboot after an update and afterwards drive was unresponsive, but still showed up in BIOS and OS. Mulitiple attempts of unplugging/powering off did not help this time. So, I was forced to do a full format and lost ~2 TiB of logging from ASU (I have ASU data from Friday though). I now have to re-copy static data over and start test back up, but here is a minor update...

868.82 TiB (955.28 TB) total.
~2224 hrs (Torture), 3257 hrs (Power-On).
16896 Raw Wear.
117.61 MB/s avg for the last 3.63 hours (on W7 x64).
MD5 every 20 loops on 1.59GB file = OK.
C4-Erase Failure Block Count (Realloc Sectors) at 24.
C5-Read Failure Block Count (uncorrectable bit errors) from 7 to 15.
1=Bnk 6/Blk 2406 - Erase Failure C4
2=Bnk 3/Blk 3925 - Erase Failure C4
3=Bnk 0/Blk 1766 - Erase Failure C4
4=Bnk 0/Blk 829 - Erase Failure C4
5=Bnk 4/Blk 3191 - Erase Failure C4
6=Bnk 7/Blk 937 - Erase Failure C4
7=Bnk 7/Blk 1980 - Erase Failure C4
8=Bnk 7/Blk 442 - Erase Failure C4
9=Bnk 7/Blk 700 - Erase Failure C4
10=Bnk 2/Blk 1066 - Erase Failure C4
11=Bnk 7/Blck 85 - Erase Failure C4
12=Bnk 4/Blk 3192 - Erase Failure C4
13=Bnk 7/Blk 280 - Erase Failure C4
14=Bnk 3/Blk 2375 - Erase Failure C4
15=Bnk 7/Blk 768 - Erase Failure C4
16=Bnk 7/Blk 765 - Erase Failure C4
17=Bnk 7/Blk 182 - Erase Failure C4
18=Bnk 5/Blk 939 - Read Failure C5
19=Bnk 5/Blk 1115 - Read Failure C5
20=Bnk 5/Blk 1011 - Read Failure C5
21=Bnk 7/Blk 3549 - Read Failure C5
22=Bnk 7/Blk 3556 - Read Failure C5
23=Bnk 4/Blk 1961 - Erase Failure C4
24=Bnk 7/Blk 1862 - Erase Failure C4
25=Bnk 7/Blk 111 - Erase Failure C4
26=Bnk 5/Blk 902 - Read Failure C5
27=Bnk 7/Blk 560 - Erase Failure C4
28=Bnk 7/Blk 827 - Erase Failure C4
29=Bnk 6/Blk 482 - Erase Failure C4
30=Bnk 5/Blk 167 - Read Failure C5
31=Bnk 7/Blk 1771 - Erase Failure C4
32=Bnk 5/Blk 956 - Read Failure C5
33=Bnk 5/Blk 1242 - Read Failure C5
34=Bnk 5/Blk 1461 - Read Failure C5
35=Bnk 5/Blk 1800 - Read Failure C5
36=Bnk 5/Blk 1030 - Read Failure C5
37=Bnk 5/Blk 1074 - Read Failure C5
38=Bnk 5/Blk 1128 - Read Failure C5
39=Bnk 5/Blk 1427 - Read Failure C5

So a total (B8+C3+C4+C5) of 49 Bad Blocks out of 32768.

[B.A.T]

I'm ready if you are Christopher. I'll post rest of the info in 10 min.

[Christopher]

I think I'm ready. I'll wait for you to post back before starting.

[B.A.T]

Todays update:
Kingston V+100
280.3839 TiB
1252 hours
Avg speed 36.55 MiB/s
AD still 1.
168= 1 (SATA PHY Error Count)
P/E?
MD5 OK.
Reallocated sectors : 00


Intel X25-M G1 80GB
97,2018 TiB
19386 hours
Reallocated sectors : 00
MWI=71 to 71
MD5 =OK
44.45 MiB/s on avg


And here is Crucial m4 64GB take II
Some smartinfo:


After I filled it with static data:


And ASU benchmark

[Christopher]

Alright then? Ready?

Are you really using 39.3GB of static data (or are you running ASU)? The Samsung is using 18.5GB with 41.0GB free space. I ran some benches and copied 18.5GB of static data to the drive, so before testing started I had 23GB shown in smart data.

EDIT
whoops, that would be the temp sensor. Most of the attributes are Vendor Specific of Unknown in CMI.


This thing is very, very fast (for the first couple of loops, but then it hits a wall). Faster than the Mushkin when running 46 percent data...

I may need to change the end of loop delay back up to 10 seconds.

[B.A.T]

I'have been using 39,3 static on both the Kingston and m4. Ready, set, go!!

Edit:
btw how is the speed in ASU? Here is mine

[Christopher]

Does anyone know what is going on here?





B.A.T.
The beginning of the loop starts at 153MiB/s, and at the end drops to 125MiB/s. After 11 loops the Avg MB/s is 114.44

[bluestang]

FYI...

CrystalDiskInfo 4.1.4 (12/04/2011) is out.

SSD-Life 2.2.39 (11/29/2011) is out.

[Christopher]

I just tried multiplying the F1 SMART value by 512 then dividing by 1024^3, and I think it's a write sector count. I'll keep watching it.

527GiB written in ASU + copying static data, etc =

542.83 if I multiply F1 * 512 then divide by 1024^3. That works out pretty well to what I wrote to the drive before I started testing.

----------------------------------------
Maybe this is common knowledge? You would think the Samsung Magician software would have the ability to show SMART attributes in a useful way, but alas, it does not.

I couldn't find anyone make mention of this, but it looks like an Indilinx write sector count, and lo and behold! it's pretty close.

I'm using the latest portable installation of SSDlife I could find, and you can see it in a screen shot above. CDI doesn't know what the hell is going on either.

The Samsung Magician software is pretty damn slick. It has everything BUT SMART viewing, including built in benchmarks and Intel like optimization of the drive and system SSD tuning, intergration with Norton Ghost for cloning and backup, FW updates, and some other random stuff.
-----------------------------------------

The chassis design is very classy, and the retail package is quite nice as well. Someone spent a lot of time designing the whole package. It kinda felt like opening the world's most interesting Christmas present.

[B.A.T]

With those speed this match is soon over. You got over 40 MiB/s faster avg writespeed than me.
The Samsung 830 looks like a very good ssd
Wonder what the WA will be?

[Christopher]

With those speed this match is soon over. You got over 40 MiB/s faster avg writespeed than me.
The Samsung 830 looks like a very good ssd
Wonder what the WA will be?

It's... fast. Really fast in a straight line. Not sure how well it does in the corners.

I agree, the match could be over soon... but only because this thing writes too fast and dies in a fiery explosion.

We'll have to see whether it can maintain this pace or not. The 64GB Samsung is roughly as fast as two Intel X25-M 80s in RAID0. That's actually pretty impressive.


It seems to have stopped doing the random section... random writes have not increased in an hour.

[bluestang]

It seems to have stopped doing the random section... random writes have not increased in an hour.

Mine did that on occasion. Would hit so many loops and then stop RW's. Has been fine for about 3 wks now though.

[Christopher]

I can't find any info about attribute F1/241 on the Samsung, but if it's not a write counter I don't know what is.

Since starting the endurance test I've logged 1345.22GiB.

The F1 value now equals 2876986359.

( 2876986359 * 512 ) / 1024^3 = 1371.84GiB.

I wrote 18.5GB of static data to the drive, and ran two quick benchmarks.

Voila! Write counter.

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

The latest version of SSDlife was supposed to have support for the 830s (you can see the shot above). I'm making inquiries.