Formatting the Seagate 1.5tb

C

Copyright

Supreme [H]ardness
Joined
Nov 17, 2000
Messages
6,004
I am reading a lot of ppl are saying it takes almost 4 hours to format this drive. I assume they are not using an NTFS quick format. Should I not use the quick format also?
 
I usually do a full format as opposed to a quick format on a new drive because the full format may cause the drive to reallocate bad sectors that could cause you trouble in the future if you only do a quick format.
 
A quick format will not check to see that all of those newly created sectors are actually readable. It's your data.

I just ordered 3 brand new WD 1 TB GP drives to go in my new NAS. 1 of the 3 failed during the initial Raid 5 array creation. It is currently in the ZZF RMA process and due in on Monday.

Just because it is a new drive does NOT guarantee that it is any good.

Start the format, go to bed, and it should be done by the time you wake up.

Don
 
It's your 4 hours, do a full format is you want. But doing it to find bad sectors is ludicrous. The OS / hardware is going to find them when your first try to write data to them (which is all a full format does anyway). Besides, that's what backups are for right?
 
ryan_975 said:
It's your 4 hours, do a full format is you want. But doing it to find bad sectors is ludicrous. The OS / hardware is going to find them when your first try to write data to them (which is all a full format does anyway). Besides, that's what backups are for right?
Click to expand...

I was just poking around Seagates website earlier today for other reasons and noticed that they listed the return rate for their current drives at .34%.

Now, they ship allot of drives. But, if you want to find out if you are 1 in 300 now, or 6 months from now, help yourself. It's your data. The drive will never fail at the worst possible time, no, that never happens.

Finding a few bad sectors is not the issue. Discovering that you have a useless piece of crap hard drive (like I did) is the issue. Most likely it is good. But.......

How important is your data?

Don
 
DonDon said:
I was just poking around Seagates website earlier today for other reasons and noticed that they listed the return rate for their current drives at .34%.

Now, they ship allot of drives. But, if you want to find out if you are 1 in 300 now, or 6 months from now, help yourself. It's your data. The drive will never fail at the worst possible time, no, that never happens.

Finding a few bad sectors is not the issue. Discovering that you have a useless piece of crap hard drive (like I did) is the issue. Most likely it is good. But.......

How important is your data?

Don
Click to expand...

My data is important enough to have current backups of it and not rely on the quality control of a manufacturer shipping out millions of drives a month.

How does formatting discover a useless drive anymore than trying to install your OS or copying your por...err data to it would?
 
ryan_975 said:
My data is important enough to have current backups of it and not rely on the quality control of a manufacturer shipping out millions of drives a month.

How does formatting discover a useless drive anymore than trying to install your OS or copying your por...err data to it would?
Click to expand...

Just 2 weeks ago I purchased 3, 1 TB WD GP drives to go into my new Thecus N5200 NAS. During the initial format of the Raid 5 in the NAS, 1 of the 3 drives failed with multiple SMART errors. That was several hours into the format. Would a quick format of that drive uncovered the problem? Prolly not.

The replacement drive is due in on Monday. I hope the new drive survives the initial LONG format that my RAID controller does. I would do nothing less.

Whats a few hours when the device may be called upon to run for the next 3 to 5 (or longer) years?

Do the long format.

Look at my sig. My current NAS is running as RAID 1. I still keep 2 copies of my important data. I have had to deal with multiple backup failures in the past. It aint pretty.

Don
 
Usually if a drive is going to fail it's going to fail soon after you start using it. A full format is a decent first test to make sure you don't have a bad drive.
 
for a drive that big i will never put my data on it, unless i stressed the trive for a while first. do some benchmarking, full drive scans with seatools an the like...

I had a 1tb drive die on me during my Raid5 tests so I was lucky it failed befor the data wad on it...
 
If you want to fully test the drive and not take hours and hours to do your XP install, use quick format then run Seagate Tools after the computer is setup. It can run all night, run in the background, etc etc. It is also much more thorough and long than a full format. If I recall, it took over 12 hours to run the extended test on just a 500GB hard drive.

^^ greater than relying on a full format if you want to be safe AND not take hours for initial XP install.
 
Thx for the info.. drives shows up tomorrow and I plan on putting it through some hell first. This is just a backup drive that will be backing up data I have across three other drives. It will probably stay unplugged accept for when im ready to do backups.
 
for some reason i don't trust these larger drives, i dont know why.. i had one Seagate 750 go bad on me, then after a check it was fine, then went bad again, ever since then i haven't trusted anything over 500G with data i care about....
 
MrGuvernment said:
for some reason i don't trust these larger drives, i dont know why.. i had one Seagate 750 go bad on me, then after a check it was fine, then went bad again, ever since then i haven't trusted anything over 500G with data i care about....
Click to expand...

I have two 750 7200.11 drives that have been great. If it dies it dies. Thats why it has a 5 year warranty and thats why I have my data backed up.
 
I agree.

I guess it would just be nice this day in age with how far technology has come for people to not have to worry so much about backups.home raid arrays...

Shouldn't computing become easier for the home user, it seems more and more our homes are turning into server rooms with 1 system for this, one for that, one for backing up this, one backup for the back up, et cetera!
 
I like these huge drives primarily for back ups - that's how I'm using my 1.5TB Seagate. I prefer just one huge external drive that I update often rather than spanning across multiple smaller drives. Then for really important data I'll make a second or third backup using older hard drives that I've retired from use.

My point is that it doesn't seem very rational to avoid large drives for your data when you can also have large drives for your backups.
 
MrGuvernment said:
I agree.

I guess it would just be nice this day in age with how far technology has come for people to not have to worry so much about backups.home raid arrays...

Shouldn't computing become easier for the home user, it seems more and more our homes are turning into server rooms with 1 system for this, one for that, one for backing up this, one backup for the back up, et cetera!
Click to expand...

Look, I am a few years older than you. The mere fact that you can view, let alone backup your media is a huge improvement over when I was younger. The ability to play any song from my music collection at any time is an amazing advance over from when I was a kid.

!!! BACK IN MY DAY WARNING!!!

Remember, the year you were born, the CD (and the DVD) did not exist. MP3 and MPG were not words yet. Hard drives for home computers did not exist. The PC, as you know it, had not been invented yet. The vinyl LP was still the only way to get HIFI audio playback. The rich were the only people with VCRs.

Despite all the hoops we must jump through to ensure the integrity of our data, it is much better and safer than in the relatively recent past.

Don
 
I would just do a quick format and be done with it.
 
SockMan! said:
I like these huge drives primarily for back ups - that's how I'm using my 1.5TB Seagate. I prefer just one huge external drive that I update often rather than spanning across multiple smaller drives. Then for really important data I'll make a second or third backup using older hard drives that I've retired from use.

My point is that it doesn't seem very rational to avoid large drives for your data when you can also have large drives for your backups.
Click to expand...

Very reasonable and the CORRECT logic. The fact is it doesn't matter if you lose 1 byte or a Gigabyte if the data isn't backed up. The fact is all HDD's MUST eventually fail (second law of thermodynamics). The fact is the more you have of an item, the higher the chance that one of those items will fail that year.
 
DonDon said:
Now, they ship allot of drives. But, if you want to find out if you are 1 in 300 now, or 6 months from now, help yourself. It's your data. The drive will never fail at the worst possible time, no, that never happens.
Click to expand...
Doing a full format doesn't guarantee a sector never goes bad, or is always writable or readable. If you do a full format, you might find that it's writable now. But six months from now, the same sector might have deteriorated.
 
Took over 4 hours to format. I ran a smart test and it passed. I know SMART isnt very dependable anyways. I plan on doing all the tests seatools has and then ill start using the drive. So far so good.
 
Id say I haven't done a complete format of a new drive in 10 years, (just a quick) and never once had a problem with bad sectors from the get-go.

Ive got a 1.5TB in the mail, and I was planning on quick-formating it.
 
Please note that when you format a drive in windows(format), Linux(fdisk), solaris(format) it is doing absolutely nothing to the drives sectors/blocks. All it is doing is adding the file system to the drive which includes the superblock and inodes for drive use.|

If you want to reset the sector/block size, you need a low level format. In todays world you can do this but only on certain drives. On very large drives, the software needed to map to the solution is just way to complex for a home user to do, though it can be done at times. Its just very complex and very long and tedious. All drives are now pre-formatted at the low level by the manufacturer.
 
Scotch77 said:
Please note that when you format a drive in windows(format), Linux(fdisk), solaris(format) it is doing absolutely nothing to the drives sectors/blocks. All it is doing is adding the file system to the drive which includes the superblock and inodes for drive use.|

If you want to reset the sector/block size, you need a low level format. In todays world you can do this but only on certain drives. On very large drives, the software needed to map to the solution is just way to complex for a home user to do, though it can be done at times. Its just very complex and very long and tedious. All drives are now pre-formatted at the low level by the manufacturer.
Click to expand...

A quick format just writes the filesystem information, a full format zeros the 512 bytes per sector that contains actual data. No one said anything about hardware level (i.e. low level) formatting. Some people do refer to a full format as a low level format, but they aren't meaning what you described.
 
Low level formatting used to be required to map the bad sectors on the drive. Sometimes if the drive developed new bad sectors, you could redo the low level format to map around the new bad blocks. It usually meant your drive was on it's last legs, so for most people it was a waste of their time.

Modern drives detect and remap the bad sectors on the fly. There is a set number of spare sectors on your drive. Once the drive starts to fail though, the spare sectors get used up pretty quick. I think their is a SMART parameter or 2 that tracks this info.

Early IDE drives were not low level formattable (is that a word?) in the field, although some manufacturers added that capability to their diagnostic programs. I just considered the drive toast though if it ever got to that level.

Don
 
DonDon said:
Low level formatting used to be required to map the bad sectors on the drive. Sometimes if the drive developed new bad sectors, you could redo the low level format to map around the new bad blocks. It usually meant your drive was on it's last legs, so for most people it was a waste of their time.

Modern drives detect and remap the bad sectors on the fly. There is a set number of spare sectors on your drive. Once the drive starts to fail though, the spare sectors get used up pretty quick. I think their is a SMART parameter or 2 that tracks this info.

Early IDE drives were not low level formattable (is that a word?) in the field, although some manufacturers added that capability to their diagnostic programs. I just considered the drive toast though if it ever got to that level.

Don
Click to expand...

I seem to remember an old ESDI drive I had a looong time ago that we had to low-level format because the controller went bad and the new one couldn't read the old one's format information.
 
I guess we could a full format. Perhaps send a couple years to do some real stress testing.

But ...

Hard drives are really cheap. Better off to by 2. Use RAID 0 (or is it 1) to duplicate the data. Toss the first one that fails.

(4 hours of my time bills out well over the cost of a 1.5GB hard drive.)
 
I've heard that you should do a full format on a new drive... and a quick format on a drive you need to reformat that you know has been working.
 
I did full format on my drives and in the middle of one, the drive started clicking. Glad I found out before I started dumping data on it.
 
amaru said:
Although it takes a long time, run the maintenance option on Spinrite
www.grc.com
Click to expand...

If anyone cares about data, although rather pricey for some of the cheapskates around here, Spinrite will save you many, many hours of frustration and prevent more dollars lost then the cost of the program.

I lvl5 all my new drives with spinrite, and in the last year it has found 2 that have serious errors 'fresh out of the box'. I used to be blind to this type of thing and because HD's are so cheap now just go buy another $80 drive, but by doing that you are only promoting this 'cheap as possible while mostly stable' crap that all the manufacturers are pushing out in the last few years.
 
mikeblas said:
Drives can fail or develop bad sectors at any time.
Click to expand...

So, are you suggesting that you should do a full format each time you format, just to be sure?

If you have the option of doing a Full or a Quick... when should you do each?
 
I'm saying that it doesn't matter. The next write could fail, whether you've previously done a quick format or a full format.

If doing a full format makes you feel better, go right ahead. It's only a few more minutes.
 
DarkStryke said:
If anyone cares about data, although rather pricey for some of the cheapskates around here, Spinrite will save you many, many hours of frustration and prevent more dollars lost then the cost of the program.

I lvl5 all my new drives with spinrite, and in the last year it has found 2 that have serious errors 'fresh out of the box'. I used to be blind to this type of thing and because HD's are so cheap now just go buy another $80 drive, but by doing that you are only promoting this 'cheap as possible while mostly stable' crap that all the manufacturers are pushing out in the last few years.
Click to expand...

The lvl5 - 'preventative maintenance' if you will - sounds like a good idea. (I do need a dedicated, crappy box to do this in though).

A quick Google search found this though:
http://209.85.173.132/search?q=cach...ite&hl=en&ct=clnk&cd=1&gl=us&client=firefox-a

Not sure how much truth it has, considering it's just random chatter though but:
<0> What it does is this: typically it'll take a sector of data, read it into RAM, then erase the sector on the drive, then write the data back to the drive and read it to verify it... then do that again, and again, upwards of 15K times per sector, one reason it takes so long
<0> But in the end, you get a drive with sectors that have been refreshed several thousand times over and each one is verified with every rewrite before the next one is worked on
<0> So when SpinRite marks a sector as bad, it truly *is* bad and useless
Click to expand...

Is there any chance of extra wear and tear on a HD, if a sector is getting written to 15K times?
 
You must log in or register to reply here.
Back
Top