Anyone with true HDDs still around here?

[RJARRRPCGP]

I've got 3 remaining 1TB HDD's from WD that I use for storage in my PC I did have 4 until the repurposed 2TB Apple HDD (Seagate 2TB with a [Cr)apple firmware installed) suddenly decided it was not going to work right anymore
Looking at it's smart readings it had somehow reached 60 degrees celsius and cooked something as it took 4.5 days just to copy off 365GB of Audio books off of it
pretty shitty of it if you ask me seeing as none of the HDD's around it got as high as 45c

I saw an HDD reach 60 C before, IIRC, and not seem to damage anything. Sure, if it was for super long periods, I expect it to have a short life, especially if that's the intake air temp!

But I have an HDD somewhere where I suspect there's a faulty sensor. And somehow, the HDD is not nearly slow as that!

Your bad HDD, likely has a PCB-level fault. I saw a Seagate with the SMART going bonkers! It's in the other room. The ECC error rate attribute, keeps going up and down!

 

[cvaldes]

I've had computers with HDDs since the Nineties and there have been HDDs in my house ever since.

Until the day something else beats them at the storage capacity-per-dollar metric, I will keep buying HDDs for media archiving (raw video recordings, photos, audio, movies/TV shows, games, etc.) and on-site backups.

Without going through the arithmetic, I figure I currently have about 80TB of HDD storage spread through a bunch of drives (mostly 8TB, 12TB, 16TB, and 18TB spinners). As time goes by, I consolidate the contents of smaller capacity drives on a larger drive and retire the old HDDs.

A couple of the drives are used for media playback but most of them are offline.

 

[Athlonite]

I saw an HDD reach 60 C before, IIRC, and not seem to damage anything. Sure, if it was for super long periods, I expect it to have a short life, especially if that's the intake air temp!

But I have an HDD somewhere where I suspect there's a faulty sensor. And somehow, the HDD is not nearly slow as that!

Your bad HDD, likely has a PCB-level fault. I saw a Seagate with the SMART going bonkers! It's in the other room. The ECC error rate attribute, keeps going up and down!

It was so bad that my PC would take over a minute to boot and as soon as I removed it back to 14 second boot times and there was no reason for it to register 60c when none of the HDD's around it barely get to 45c when being used considering it spent the majority of it's time idle I'm still of the mind that it has something to do with the [Cr]apple FW telling it to self immolate when x hours of power on time (built in obsolescence) were reached it had worked flawlessly for the last 7~8 years seatools would fail it on both long and short tests but atleast the data I got off it was fine it just took forever to to get it off onto another HDD

 

[Bagerklestyne]

I've had computers with HDDs since the Nineties and there have been HDDs in my house ever since.

Until the day something else beats them at storage-per-dollar metric, I will keep buying HDDs for media archiving (raw video recordings, photos, audio, movies/TV shows, games, etc.) and on-site backups.

Without going through the arithmetic, I figure I currently have about 80TB of HDD storage spread through a bunch of drives (mostly 8TB, 12TB, 16TB, and 18TB spinners). As time goes by, I consolidate the contents of smaller capacity drives on a larger drive and retire the old HDDs.

A couple of the drives are used for media playback but most of them are offline.

This is the same for me and at the moment the best option I could find was still 2x5 the price and lets be honest, outside of a modest power consumption reduction there needs to be a good reason, if it's in media storage or archival/backups then the i/o and transfer speeds benefits of ssd's are pretty much negated (aside from raid rebuild times perhaps)

 

[cvaldes]

This is the same for me and at the moment the best option I could find was still 2x5 the price and lets be honest, outside of a modest power consumption reduction there needs to be a good reason, if it's in media storage or archival/backups then the i/o and transfer speeds benefits of ssd's are pretty much negated (aside from raid rebuild times perhaps)

It's worth pointing out that I also have usage cases where speed in an external drive is helpful. Thus, I have an external 2TB m.2 SSD as well as an 8TB 2.5" SATA SSD in addition to my piles of 3.5" HDDs.

I use these two SSDs as necessary but not for long-term storage. However there are certainly instances where sneakernetting >1TB files is better than using the network. And yes, external SSDs have long been used as scratch disks for media editing tasks. Just because I mostly rely on HDDs for long-term storage doesn't mean I don't find SSDs useful.

 

[windwhirl]

I still have three HDDs in my main system, as shown in my system specs page:
Toshiba DT01ACA200 2TB drive
Seagate IronWolf 10 TB drive
Seagate IronWolf 8 TB drive

Lots of videos, images, backups, etc. Capacity per dollar is the winner metric for this use case, and as long as the drives are properly defragmented, they can reach their top speeds (180-210 MB/s) with the larger files. For smaller files, speed goes down the smaller they are, but that happens on SSDs as well, so it's not like it changes anything for me

 

[RJARRRPCGP]

It was so bad that my PC would take over a minute to boot and as soon as I removed it back to 14 second boot times and there was no reason for it to register 60c when none of the HDD's around it barely get to 45c when being used considering it spent the majority of it's time idle I'm still of the mind that it has something to do with the [Cr]apple FW telling it to self immolate when x hours of power on time (built in obsolescence) were reached it had worked flawlessly for the last 7~8 years seatools would fail it on both long and short tests but atleast the data I got off it was fine it just took forever to to get it off onto another HDD

Topcat (over at badcaps. net) would suspect the PCB on the HDD, needs cleaning off of the oxidation.

 

[starfals]

Me me me I got 2X 20TB disks from WD, and like 5X8TB ones. 7-8 USB ones with the size of 4 and 5TBs and a couple of 4TB SSDs (just saying, in case someone thinks im full HDD mode lol. Just cus 75% of my storage is that XD)

So far, the disks that keep breaking are USB drives my passports from WD, and also a few WD blacks. They all broke down and got corrupted. Ultra star ones work great, gotta pick anything with a star in the name after all

 

[lexluthermiester]

Western Digital had a temporary outage yesterday. A lot of stuff wouldn't respond.

Bad timing then. Thanks!

 

[Bagerklestyne]

I do love the idea of having 200-300TB in a nuc sized nas format, my DS1821+ isn't exactly a univac but a smaller footprint could give me a couple of installation options that aren't currently on the table.

 

[Princess Garnet]

I'm pretty sure, those "Blue" 640GB drives you mentioned are also the ones I bought.
Last time I checked SMART on them (before retiring) :
View attachment 365285
I think I bought them in late 2008, but I might be remembering wrong...

That's them, the "AAKS" drives.

They were fairly good performing drives for the late 2000s if I remember right (of course the "premium" drives like the Black existed, and the Samsung Spinpoint was also better, but for "general" drives they were pretty good). I remember being a bit concerned the drop to 5,400 RPM with the 4 TB drives would be felt, but it just wasn't. Higher density helps, and it's why I think 7,200 RPM drives haven't truly been relevant for perhaps over five years now. I think it you want performance, you get SSDs, and if you want storage capacity, you get better value 5,xx0 RPM drives. There's always room for a middle ground, but in this case it feels like an increasingly narrow niche that just hasn't been relevant in my opinion for a while.

I just looked at mine and both labels have an October 22nd, 2008 date. That sounds right for about the time the color nomenclature was introduced since the first 640 GB ones I had that were called "Caviar SE16" would have been earlier 2008 or even late 2007.

The other pair (the Caviar labeled drives) aren't in my possession at the moment, so I can't check them, but they may (or may not) have slightly higher hours on them. Most of these hours would have been "spun down" hours anyway since I don't keep my drives running all the time, but the drive is still powered to a degree so it still counts it as power on hours. I know they were still called 6400AAKS, but the six letter/number string that followed was different.

Why is the Vcore that high at only 2.4 GHz? (almost 1.5V!) 400x6.0 is stock core speed. Should be 1.3250V or down to 1.250V, if a better batch. Even a crappy Q6600 can do a 3.2 GHz core OC @ 1.4V.

The problem is that, Kentsfield, is the meme of "Bus/Interconnect Error" WHEAs. With dud samples, that will start happening at not much past 333 MHz FSB!

400 MHz FSB would be an overclock, no? The 65nm Core 2s had a 1,066 MHz quad FSB (266 MHz actual) and the 45nm refresh Core 2s had a 1,333 MHz FSB (333 MHz actual).

As for the second part, all of the Core 2 Quads were a pair of Core 2 Duo dies on a package that communicated over the FSB. Might have something to do with that.

 

[natr0n]

8X1TB RAID 0 - The controller has many cache settings but I disable for true speeds.
This is with Toshiba 2.5 drives. I have to find my 3.5 ones.

 

[lexluthermiester]

8X1TB RAID 0 - The controller has many cache settings but I disable for true speeds.
This is with Toshiba 2.5 drives. I have to find my 3.5 ones.
View attachment 365365

RAID0 with 8 drives! You are very ballzy sir! I would, at least, RAID5 those drives if not opting for RAID6.

 

[natr0n]

RAID0 with 8 drives! You are very ballzy sir! I would, at least, RAID5 those drives if not opting for RAID6.

it's for fun mostly. I dont put important files on there.

I do want to play with an os on it.

 

[RJARRRPCGP]

Me me me I got 2X 20TB disks from WD, and like 5X8TB ones. 7-8 USB ones with the size of 4 and 5TBs and a couple of 4TB SSDs (just saying, in case someone thinks im full HDD mode lol. Just cus 75% of my storage is that XD)

So far, the disks that keep breaking are USB drives my passports from WD, and also a few WD blacks. They all broke down and got corrupted. Ultra star ones work great, gotta pick anything with a star in the name after all

My Western Digital Blacks from the late-2010s, seem fine. I can't see why I would get corrupted data, based on tests that I ran. I deeply scanned them. Even my Western Digital Caviar Blue 320 GB from the late '00s, (one of the last HDDs with the Caviar name) passed tests.

OTOH, I have a Seagate Barracuda ST500DM002 500 GB HDD that's sus, the SMART is going bonkers, for the ECC error rate attribute! The surface test looks like it can read and write without corruption, but definitely not sure, in this case!

That's them, the "AAKS" drives.

They were fairly good performing drives for the late 2000s if I remember right (of course the "premium" drives like the Black existed, and the Samsung Spinpoint was also better, but for "general" drives they were pretty good). I remember being a bit concerned the drop to 5,400 RPM with the 4 TB drives would be felt, but it just wasn't. Higher density helps, and it's why I think 7,200 RPM drives haven't truly been relevant for perhaps over five years now. I think it you want performance, you get SSDs, and if you want storage capacity, you get better value 5,xx0 RPM drives. There's always room for a middle ground, but in this case it feels like an increasingly narrow niche that just hasn't been relevant in my opinion for a while.

I just looked at mine and both labels have an October 22nd, 2008 date. That sounds right for about the time the color nomenclature was introduced since the first 640 GB ones I had that were called "Caviar SE16" would have been earlier 2008 or even late 2007.

The other pair (the Caviar labeled drives) aren't in my possession at the moment, so I can't check them, but they may (or may not) have slightly higher hours on them. Most of these hours would have been "spun down" hours anyway since I don't keep my drives running all the time, but the drive is still powered to a degree so it still counts it as power on hours. I know they were still called 6400AAKS, but the six letter/number string that followed was different.

400 MHz FSB would be an overclock, no? The 65nm Core 2s had a 1,066 MHz quad FSB (266 MHz actual) and the 45nm refresh Core 2s had a 1,333 MHz FSB (333 MHz actual).

As for the second part, all of the Core 2 Quads were a pair of Core 2 Duo dies on a package that communicated over the FSB. Might have something to do with that.

IIRC, the AAKS were likely from 2008-2009-2010.
They became "Blue" by 2009, which they were called "Caviar Blue" in 2009. It was by sometime after 2009, by likely 2013, if not sooner, when Western Digital discontinued the Caviar name.

 

[RaceT3ch]

Here's off-the-charts-paranoia I predict that is possible: People who are "experts" will ban RAM from being sold with computers, due to a person falsely thinking that the RAM is retaining passwords!

some people shouldn't use computers

 

[RJARRRPCGP]

some people shouldn't use computers

Yep, I see paranoia where some people highly likely already think that selling used HDDs, is illegal, even when it's legal-eagle! Just because of HIPPA. Even when lots of PCs don't have patient records!

 

[natr0n]

So I installed win10 on the raid0, although it works no issues its still a bunch of hdd. feels sluggish. Defragments in a minute.

I am going to try using it for games that might work better.

Another thing would be get a 8-10 pack of cheap ssd and have a party with no limits.

 

[Princess Garnet]

For the OC, yes, you got a nice FSB OC, better than my Q6600 for sure, but not the cores, so you can turn the Vcore down. Even my dud didn't need voltage changes for DDR2 666/667, which was very common for DDR2 in the late DDR2 days. So, I only had to touch the VDIMM when I was going to fill all the slots, for 8 GB of DDR2. For filling all the RAM slots, I used 1.90V for VDIMM.
(The BIOS would crash, if I didn't raise the VDIMM!) (freeze at the Asus P5QL Pro logo, with the Asus P5QL Pro then I used the same voltage after going to the Asus Maximus II Gene)

333x9.0 was easy-peasy, for even the dud Q6600 G0 L804. That was likely at just 1.3250V for Vcore as well!

I think you're confusing me with someone else. The overclocked Q6600 isn't mine. I never had one (I went with the E8400/E8600 instead, and it was on an X38 Maximus Formula). I was just commenting on it.

IIRC, the AAKS were likely from 2008-2009-2010.
They became "Blue" by 2009, which they were called "Caviar Blue" in 2009. It was by sometime after 2009, by likely 2013, if not sooner, when Western Digital discontinued the Caviar name.

My pair of Blue drives are October of 2008, and my Cavair drives were late 2007 to mid 2008 (I honestly forget when the two 320 GB drives were purchased and when the first pair of 640 GB drives were purchased, but the 640 GB would have been late 2007 at earliest and mid 2008 at latest), so sometime between then was the switchover it seems. Probably late 2008 to be more exact.

 

[lexluthermiester]

it's for fun mostly. I dont put important files on there.

I do want to play with an os on it.

Ah ok, gotcha!

So I installed win10 on the raid0, although it works no issues its still a bunch of hdd. feels sluggish. Defragments in a minute.

For technical reasons I'm not going to go into here, RAID 5 or 6 would work better. That may seem counter intuitive, but it works better. RAID0 for boot drives work best with only 2 or 3 drives.

I am going to try using it for games that might work better.

Very likely.

 

[natr0n]

Ah ok, gotcha!


For technical reasons I'm not going to go into here, RAID 5 or 6 would work better. That may seem counter intuitive, but it works better. RAID0 for boot drives work best with only 2 or 3 drives.


Very likely.

Thanks, I will try. I only ever used raid 0 and 1.

 

[lexluthermiester]

Thanks, I will try. I only ever used raid 0 and 1.

I'm not trying to twist your arm or anything. I'm a bit of a RAID geek. Been doing RAID arrays since the days of SCSI 50pin connectors. It's been a hot minute..

 

[RJARRRPCGP]

View attachment 365288

Why is the Vcore that high at only 2.4 GHz? (almost 1.5V!) 400x6.0 is stock core speed. Should be 1.3250V or down to 1.250V, if a better batch. Even a crappy Q6600 can do a 3.2 GHz core OC @ 1.4V.

The problem is that, Kentsfield, is the meme of "Bus/Interconnect Error" WHEAs. With dud samples, that will start happening at not much past 333 MHz FSB!

I think it's honestly because of 4 cores being an afterthought at the time of 65nm Core 2s. Windows (Windows 7) showed in the event log that it was always for APIC IDs 3 or 2, I believe. That means the 4th core or the 3rd core having a bus communication problem.

Kentsfield is the meme of needing to crank up the FSB termination voltage, when much past 333 MHz FSB, even with the P45 chipset. I started noticing that issue especially past 350 MHz FSB on my dud.

I believed I tried different GTL settings, but that it didn't make a difference like FSB termination voltage did.

Even you my dud Q6600 G0 L804, 333x9.0, was easy-peasy. With Vcore at 1.3250V.

 

[agent_x007]

Why is the Vcore that high at only 2.4 GHz? (almost 1.5V!) 400x6.0 is stock core speed. Should be 1.3250V or down to 1.250V, if a better batch. Even a crappy Q6600 can do a 3.2 GHz core OC @ 1.4V.

Because I had C1E/SpeedStep active - actual core speed was 3.6GHz
I did a lot of "silly things" back in day... and still do
About that Q6600 :


Recently/today :

 

[lexluthermiester]

Because I had C1E/SpeedStep active - actual core speed was 3.6GHz
I did a lot of "silly things" back in day... and still do
About that Q6600 :
View attachment 365464

Recently/today :
View attachment 365474

Good grief!! 4ghz?!? Damn..