| HDD GURU FORUMS http://forum.hddguru.com/ |
|
| Experiments - WD Green SSD Sandisk 20-82-000496-2 Recovery http://forum.hddguru.com/viewtopic.php?f=27&t=43423 |
Page 2 of 2 |
| Author: | Amarbir[CDR-Labs] [ June 20th, 2023, 13:46 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
melvin wrote: I meant visual inspection, not checking ECC map. It's not true that if NAND is overheated all pages would be red. I think i've red them with 3,3V VCC and VCCq. Speed high. RR worked so good that i didn't change enything. And you can see how bad it was before. Was ECC also checked and confirmed? Well, There is Not Little But Lots And Lots of progress with this case .I will be debating that IR Versus Hot Air Scenario Very Soon . But Generally how many days are you talking for 120,240,480 and 960GB cases for this SSD Average . |
|
| Author: | melvin [ June 20th, 2023, 15:28 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
It can be done within one week, depending on cells condition. |
|
| Author: | Amarbir[CDR-Labs] [ June 21st, 2023, 5:21 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
melvin wrote: It can be done within one week, depending on cells condition. Well, One week for which capacity are you talking about,From 4 BGA 132 i am now doing U9 With Near 5% Pending ECC Corrections for Whole Dump ,Do you still think i cooked The Chips  , This is nearing Arvika Percentage of Recovery  .BTW i am always very fortunate that i get attention and replies from you ,arvika and bolo sir ,All great engineers ...
|
|
| Author: | melvin [ June 21st, 2023, 7:55 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
One week for any capacity of those you asked. |
|
| Author: | Amarbir[CDR-Labs] [ June 21st, 2023, 9:12 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
melvin wrote: One week for any capacity of those you asked. Very Nice , Just Using VNR or Mixing VNR+FE For doing tasks side by side |
|
| Author: | melvin [ June 21st, 2023, 12:01 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
You can do it side by side in VNR. |
|
| Author: | Amarbir[CDR-Labs] [ June 21st, 2023, 13:40 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
melvin wrote: You can do it side by side in VNR. Well, We Can Have Another Computer and Do Tasks which do not use the reader ,But if you have to RR then you need to have two VNR's or a VNR and FE Or Similar Tool .How Good is FE RR for these Chips ? . |
|
| Author: | melvin [ June 21st, 2023, 14:57 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
You can read in one task, correct in another for example and reread overnight. Did you read the article about how to recover data from those SSDs with VNR? FE with proper RR can do magic but there's always need to read all chips at least 2-4 times. Something i don't like about FE is that when you want to reread one page, you have to read the whole chip. |
|
| Author: | arvika [ June 21st, 2023, 16:21 ] | ||
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove | ||
Amarbir[CDR-Labs] wrote: Well, We Can Have Another Computer and Do Tasks which do not use the reader ,But if you have to RR then you need to have two VNR's or a VNR and FE Or Similar Tool .How Good is FE RR for these Chips ? . Readers are not a problem. Time is the problem. When you have one case you can play many days, even weeks. Several cases with 1-2 week work per one case can disturb the work at lab.
|
|||
| Author: | Gregory [ June 23rd, 2023, 7:56 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
Maybe page addressing problem, do you have data on whole dump? or only to block 1980 and after that empty pages? |
|
| Author: | Amarbir[CDR-Labs] [ June 23rd, 2023, 14:01 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
Gregory wrote: Maybe page addressing problem, do you have data on whole dump? or only to block 1980 and after that empty pages? Well, Pardon me i cannot understand what you mean by you this comment ,For me it makes ZERO sense |
|
| Author: | Amarbir[CDR-Labs] [ June 23rd, 2023, 14:03 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
melvin wrote: You can read in one task, correct in another for example and reread overnight. Did you read the article about how to recover data from those SSDs with VNR? FE with proper RR can do magic but there's always need to read all chips at least 2-4 times. Something i don't like about FE is that when you want to reread one page, you have to read the whole chip. Well, I am about to order FE , How many readers should i order 4 are enough right ,I really wanna try it
|
|
| Author: | Amarbir[CDR-Labs] [ June 23rd, 2023, 14:06 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
arvika wrote: Amarbir[CDR-Labs] wrote: Well, We Can Have Another Computer and Do Tasks which do not use the reader ,But if you have to RR then you need to have two VNR's or a VNR and FE Or Similar Tool .How Good is FE RR for these Chips ? . Readers are not a problem. Time is the problem. When you have one case you can play many days, even weeks. Several cases with 1-2 week work per one case can disturb the work at lab. HaaHaa , This is like that cherries harvest you posted on facebook ,Bloody hell 8 FE Readers , I will be Ordering My FE Very Soon "I am thinking 4 Readers Should Be Good Enough " . BTW i am now on the second chip ,First One Now i have less then 4% errors and i am sure more corrections are possible ,I have only done 5 Passes On U9 .But there was a trick involved in achieving that in VNR ,In FE you folks play with the NAND and find what works |
|
| Author: | Amarbir[CDR-Labs] [ June 25th, 2023, 13:44 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
So Friends , Using My Trick I Was able to get to less then 1% BAD Pages Only Infact 0.8% On Chip U6 , I am Now Reading U8 And Will start ECC Correction And Readretry " RR " Using My Little Trick For This ID .I Will Start To Read U5 On a Different System And Will Use Another For ECC Correction ,2 Systems For 1 Job and One Reader . PS : Has Anyone Tried This Tool On a Windows 11 System ? . |
|
| Author: | melvin [ June 25th, 2023, 14:51 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
If it's 1% from the whole chip it's ok. If from the blocks with data it should be better. I don't use W11. It sucks as a tool but it's much faster than W10. |
|
| Author: | Amarbir[CDR-Labs] [ June 25th, 2023, 23:38 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
melvin wrote: If it's 1% from the whole chip it's ok. If from the blocks with data it should be better. I don't use W11. It sucks as a tool but it's much faster than W10. Well , Its 1% from Whole Chip But U9 Is About 5% ,But i am sure i can correct more of that ,Currently i am reading and dumping dumps U5 and U8 ,U5 Dump should be done in about 2 hours then i will start ecc correction before RR , U8 Dump is read and ecc correction is in progress ,Should take a few hours before i can start RR . Well i got this acer nitro 7000 series ryzen 7 and all that it supports is W11 ,Windows 10 is not in support section of Acer -> https://www.acer.com/us-en/support?sear ... an515-47;1 ,When i use " Correct dump GPU " Function and use my RTX 3050 it crashes and exits .I have updated my driver and disabled onboard GPU of ryzen 7 700 series too . |
|
| Author: | sourcerer [ July 28th, 2023, 19:31 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
@Amabir: I am not sure about this, but I have been thinking about this lately too, and I think that pages can have several bit errors in them, and it could be that the ECC was able to fix one of those bit errors (so the page got corrected), but the page might still have other bit errors in them, so it would be retried in the next run again. Could this explain the behavious you have seen? |
|
| Author: | Amarbir[CDR-Labs] [ July 29th, 2023, 3:07 ] |
| Post subject: | Re: Experiments - WD Green SSD Sandisk 20-82-000496-2 Recove |
sourcerer wrote: @Amabir: I am not sure about this, but I have been thinking about this lately too, and I think that pages can have several bit errors in them, and it could be that the ECC was able to fix one of those bit errors (so the page got corrected), but the page might still have other bit errors in them, so it would be retried in the next run again. Could this explain the behavious you have seen? Well, I have two cases one is a normal SATA SSD HDD and the other is a M.2 SATA SSD , I worked on one unit for approx 1 month and got down the bit errors in data areas of NAND Dump to under 1% in all 4 chips ,Then i got TS support to work on it and they said "Currently NOT Supported " . This Was The M.2 Types ,Now i Am Working On Normal SATA ,But these cannot be done in under 1 week ,I do not know how melvin manages it i just cannot, system is running 24x7 since then on NAND Cases .Your observation above is wrong ... |
|
| Page 2 of 2 | All times are UTC - 5 hours [ DST ] |
| Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |
|