HDD GURU FORUMS

Hi All,

A few days ago I have been scammed, ordering A400 480GB but somehow the seller send me UV400. From Crystal Disk info I got this data:

Code:

           Model : KINGSTON SUV400S37 480GB
        Firmware : S0423B0
   Serial Number : AA000000000000006307
       Disk Size : 480,1 GB (8,4/137,4/480,1/----)
     Buffer Size : Unknown
     Queue Depth : 32
    # of Sectors : 937703088
   Rotation Rate : ---- (SSD)
       Interface : Serial ATA
   Major Version : ACS-3
   Minor Version : ACS-3 Revision 4
   Transfer Mode : SATA/600 | SATA/600
  Power On Hours : 72 hours
  Power On Count : 17 count
      Host Reads : 341 GB
     Host Writes : 365 GB
     NAND Writes : 96 GB
     Temperature : 30 C (86 F)
   Health Status : Good (100 %)
        Features : S.M.A.R.T., APM, 48bit LBA, NCQ, TRIM
       APM Level : 0000h [OFF]
       AAM Level : ----

So I am suspecting if this is counterfeit SSD. After googling for a while I got a clue if this SSD is using SM controller:

Code:

v0.555a
Drive: 0(ATA)
OS: 10.0 build 18362
Model: KINGSTON SUV400S37 480GB               
Fw   : S0423B0
Size : 457862 MB
From smart : [SMI2259XT] [S0423B0 00]
Controller : SM2259 
FlashID: 0x89,0xd4,0xc,0x32,0xaa,0x0,0x0,0x0 - Intel 64L(N18A) QLC 1024Gb/CE 1024Gb/die
Channel: 4
CE     : 1
TotDie : 4
Plane  : 4
Die/Ce : 1
Ch map : 0x0F
CE map : 0x01
Inter. : 1
First Fblock  :    2
Total Fblock  :  736
Total Hblock  : 7155
Fblock Per Ce :  736
Fblock Per Die:  736
Original Spare Block Count :    65
Vendor Marked Bad Block    :     0
Bad Block From Pretest     :    34


I found also SMI Mass Production Tool and seems this tragedy (sigh) opening a new world to me so I am decided to letting go the seller.







Can anyone give the pointer what is this SSD based on? and is 480GB space is real?

Thank you,

The controller is definitely from Silicon Motion (usually abbreviated as SM or SMI), I would suggest to use ChipGenius (be careful not get malware when you download it!) to analyze it, it should be able to tell you which chips are in there and which size they really have. Alternatively you could also use the Mass-Production tools from SiliconMotion, they also display the real chip details when attached.

Thanks for the reply.

BTW from the SMART I found: Cumulative ECC Bit Correction Count (Code C3). A bit googling and this value seems specific from Micron SMART, it said: This attribute provides data that is tracked by Micron engineering. It is not indicative of SSD wear or of impending failure.

Perhaps this SSD is made by home industry and that's cool. It would be great if we can creating our own SSD? perhaps using some kind of this https://www.aliexpress.com/item/33004090439.html

This kind of looks like those boards you get from Aliexpress and have to add your own chips

has this "look" about it
https://www.aliexpress.com/item/33004090439.html?spm=a2g0o.productlist.0.0.35983591iUTXZW&algo_pvid=42617be8-76b1-4d94-9f03-712f040f9de7&algo_expid=42617be8-76b1-4d94-9f03-712f040f9de7-0&btsid=19eac4cd-deb0-4e37-8925-0e2fabdf8d07&ws_ab_test=searchweb0_0,searchweb201602_2,searchweb201603_55

sourcerer wrote:The controller is definitely from Silicon Motion (usually abbreviated as SM or SMI), I would suggest to use ChipGenius (be careful not get malware when you download it!) to analyze it, it should be able to tell you which chips are in there and which size they really have. Alternatively you could also use the Mass-Production tools from SiliconMotion, they also display the real chip details when attached.

Seems I am not yet permitted to post any link

I see Cumulative ECC Bit Correction on the SMART and its specific for Micro SMART: This attribute provides data that is tracked by Micron engineering. It is not indicative of SSD wear or of impending failure.

HaQue wrote:This kind of looks like those boards you get from Aliexpress and have to add your own chips

Ya but a little different on the center but my SSD is made by home industry, cool. It would be great if we can creating our own SSD

Let me know if you know some tuts on how to do it.

When looking at the Ads, I assume the following:

1. you buy the board from Ali Express
2. you buy some chips from same, or Alibaba or one of the many chip retailers that look dodgy as eff, the ones you find when searching for datasheets... or wherever
3. you solder on chips
4. you use the mass production tools to marry the memory and setup the SSD in relation to whatever features, security, partitioning etc
5. you rejoice in using your likely slow, unreliable SSD!

the part I dont know is if the MP tools will be provided. I may have to buy one just see what it is about

I have a heap of kingstons here, I will have a look whats in them. I do know a few I looked at had some SM controllers, but not sure if it was those or not. I dont remember any of the boards I looked at being so sparce, or homebrew looking as yours

HaQue wrote:5. you rejoice in using your likely slow, unreliable SSD!

But its fun so it worth it

As long as we put all things right perhaps not that bad. Thanks anyway

do you have pictures of the case?

This is one of my Kingston's (in the usual status as Failed!) I know it is a different capacity, but usually they are very similar between capacities of the same model.

Note the UV400 stamp on inside of cover, marvell controller and TSOP48 memory chips rebranded as Kingston.

I already put it inside my old laptop and run stable so far, noticed decreased performance compared to previous SSD Kingston v300 128GB (not fake), because it has slower random IOPS.

A few pic from days ago:







The Kingston SSD Manager also can't detect it so its 100% fake. Curious how long it will last but from current performance I think it will fine for years.

The SMART from various tools also looks fine:

Code:

[Self-Monitoring, Analysis and Reporting Technology (S.M.A.R.T.)]
  [01] Raw Read Error Rate:               100/50, Worst: 100
  [05] Reallocated Sector Count:          100/50, Worst: 100
  [09] Power-On Hours/Cycle Count:        100/50, Worst: 100 (89 hours / 3.7 days)
  [0C] Power Cycle Count:                 100/50, Worst: 100 (Data = 18,0)
  [A0] Uncorrectable Sector Count:        100/50, Worst: 100
  [A1] Valid Spare Blocks:                100/50, Worst: 100 (Data = 100,0)
  [A3] Initial Invalid Blocks:            100/50, Worst: 100 (Data = 34,0)
  [A4] Total TLC Erase Count:             100/50, Worst: 100 (Data = 2421,0)
  [A5] Maximum TLC Erase Count:           100/50, Worst: 100 (Data = 8,0)
  [A6] Minimum TLC Erase Count:           100/50, Worst: 100 (Data = 1,0)
  [A7] Average TLC Erase Count:           100/50, Worst: 100 (Data = 3,0)
  [A8] SATA PHY Error Count:              100/50, Worst: 100 (Data = 1500,0)
  [A9] Percentage Lifetime Remaining:     100/50, Worst: 100 (Data = 100,0)
  [AF] Program/ECC Fail Count:            100/50, Worst: 100
  [B0] Erase Fail Count:                  100/50, Worst: 100
  [B1] Wear Leveling Count:               100/50, Worst: 100
  [B2] Used Reserved Block Count (Chip):  100/50, Worst: 100
  [B5] Program Fail Count (Total):        100/50, Worst: 100
  [B6] Erase Fail Count (Total):          100/50, Worst: 100
  [C0] Power-off Retract Count:           100/50, Worst: 100 (Data = 3,0)
  [C2] Temperature                        100/50, Worst: 100 (30.0 °C)
  [C3] Cumulative ECC Bit Correction:     100/50, Worst: 100 (Data = 28,0)
  [C4] Reallocation Event Count:          100/50, Worst: 100
  [C5] Current Pending Sector Count:      100/50, Worst: 100
  [C6] Off-Line Uncorrectable Sector:     100/50, Worst: 100
  [C7] SATA CRC Error Count:              100/50, Worst: 100
  [E8] Available Reserved Space:          100/50, Worst: 100 (Data = 100,0)
  [F1] Total Host Writes:                 100/50, Worst: 100 (Data = 12202,0)
  [F2] Total Host Reads:                  100/50, Worst: 100 (Data = 13625,0)
  [F5] Total TLC Write Count:             100/50, Worst: 100 (Data = 3488,0)
  Drive Remaining Life                    100%
[Device Statistics]
  Lifetime Power-On Resets:               18
  Power-on Hours:                         89
  Logical Sectors Written:                799713743
  Logical Sectors Read:                   892959177
  Number of Write Commands:               26290036
  Number of Read Commands:                24819752
  Used Endurance Indicator:               0%

I think the chips are fake as well as it looks like a poor attempt at a Micron logo, plus I think a 1 Tb chip should be numbers like 29F1T08 not 29F1TB

I would test the disk by writing it fully and testing back to make sure of no fake capacity shenanigans like storing a bit of data and getting full on the real memory capacity but simply writing over itself.

tools like H2testw can do it. some here (virus scan them first!) : https://www.raymond.cc/blog/how-to-check-and-test-usb-flash-drive/

HaQue wrote:I think the chips are fake as well as it looks like a poor attempt at a Micron logo, plus I think a 1 Tb chip should be numbers like 29F1T08 not 29F1TB

I would test the disk by writing it fully and testing back to make sure of no fake capacity shenanigans like storing a bit of data and getting full on the real memory capacity but simply writing over itself.

tools like H2testw can do it. some here (virus scan them first!) : https://www.raymond.cc/blog/how-to-check-and-test-usb-flash-drive/

OK this is scare me actually. By filling all of the disk it will speed up the wearing level it seems. I do not know how flash work to saving the data, let say I have flash memory 1TB per chip, so technically can we have 4TB disk space on it?

Thanks, but testing like that will increase the wear leveling, I hope its real 480GB

AS SSD Benchmark:



I make the SSD as Windows partitions (have system image for it). I am creating 2 partitions on it, first around 360GB (C) and the rest for second partitions (D) and I put data on second partitions and its fine.

4TB??? No! the 1Tb chips mean 1 terabit not terabyte. 8 bits make a byte. depending on the exact bytes on the chip it is roughly about right for a 480GB SSD.


No need to be scared, you are filling disk 1 time. flash cells are fine to program tens of thousands. SLC around 10,000, 3D NAND 35,000

from WD:

SSD endurance is commonly described in terms of Drive Writes Per Day (DWPD) for a certain warranty period (typically 3 or 5 years). In other words, if a 1TB SSD is specified for 1 DWPD, it can withstand 1TB of data written to it every day for the warranty period.

I dont think you want to use a known fake disk for data until you know it is full capacity. So far you might be OK because the chips may be 1Tb each.


example:
a fake 64GB flash drive may contain a fake NAND chip of 8GB. so you go do some testing and write a heap of files on. all looks ok, apart from the dog awful write speed.
you might write 4GB, 6GB all is ok so you think great I got a 64GB drive for 10 bucks!

later you write another few GB and go to around 9GB and find now the data is overwriting on the first 1GB again. They can write firmware so the drive works like this, but wont throw errors. Pretty clever.

FWIW, the OP ordered an A400 480GB. The real Kingston A400 uses a Phison PS3111-S11 DRAM-less controller, at least in the 120GB and 240GB versions.

https://www.hardwaresecrets.com/kingston-a400-120-gib-ssd-review/
https://www.digit.in/reviews/pc-components/kingston-a400-ssd-240-gb-review-125099.html

The fake SSD is also DRAM-less, albeit with a different controller.

HaQue wrote:later you write another few GB and go to around 9GB and find now the data is overwriting on the first 1GB again. They can write firmware so the drive works like this, but wont throw errors. Pretty clever.

This is the most bad upgrade I have done so far. I have an ancient ASUS K43 Series. Upgrade the CPU to i7-2670qm, modding the AMI BIOS so the hidden BIOS options show up (update the Intel ME and CPU microcode to the latest as well). The official specs is max to 8GB RAM but I can put 16GB right (DDR3 1333) perhaps can up to 32GB.

fzabkar wrote:FWIW, the OP ordered an A400 480GB. The real Kingston A400 uses a Phison PS3111-S11 DRAM-less controller, at least in the 120GB and 240GB versions.

Use RW everything and found:

Code:

Bus 00, Dev 1F, Fun 02
Memory Base DF006000, Port 0
Support ATA/ATAPI-10
ATA device
Model number   KINGSTON SUV400S37 480GB
Firmware revision   S0423B0
Serial number   AA000000000000006307
Total Sectors (48bit)   37E436B0 (480GB)
PIO mode      4
Multiword DMA   Selected/Supported = None/2
Ultra DMA   Selected/Supported = 5/6
Cable report   0: 40pin
SMART feature   Enabled/Supported = Yes/Yes
Supported speed   Gen3 (6G), Gen2 (3G), Gen1 (1.5G)
Interface Speed   Gen3 (6Gbps)


Googling about 37E436B0 and it from Micron M510DC / Micron 5200 series. Found it also from Crucial BX500 datasheet. Need to entering the ROM mode to get more info.

But anyway my old laptop now have a new life:

Code:

https://browser.geekbench.com/v4/cpu/14237859

The number of sectors (0x37E436B0) is an IDEMA standard.

0x37E436B0 (hexadecimal) = 937,703,088 in decimal

http://www.google.com/search?q=0x37E436B0+in+decimal

http://www.google.com/search?q=937703088

your ram is likely limited by the chipset of the mainboard.