Skip to main content

i recently rented an Alesis ADAT hd24xr to record a live performance. things went fine until the event was over, and i lost power to my workstation before i could save the files, or even stop the recorder.

long story short, i'm trying to see if anyone knows, or knows someone who might know, how to attempt to recover my lost data. again, the recorder was running fine through the entire performance, so the data has to be there in some format. i attempted to ftp in to the drive using the rj45 connector, but all i could find were the header files. if anyone knows anything about i might be able to perform a recovery, i would certainly appreciate the help.

thanks in advance...

Comments

RemyRAD Sun, 04/16/2006 - 21:04

The Alesis offers a FireWire adapter interface that allows one to remove the disk drive from the machine. Once the disk is removed and connected to the FireWire adapter, you can connect it to your computer and dump all 24 channels into your computer via the FireWire port and into a substantially sized storage hard drive within your computer. DO NOT ATTEMPT THIS WITH THAT DISK DRIVE WITHOUT THEIR PROPRIETARY FIREWIRE ADAPTER! Because, Alesis uses a proprietary disk file format that negates the use of defragmentation programs. I.e. it is not FAT 32 or NTFS as is most commonly used for IBM clones and their similar Macintosh cousins. So it is not compatible with any computer based system, " out of the box".

On that particular unit, if you were recording, it's on the disk drives. Unlike other computer-based systems that require you to " save files" at the end of recording, the Alesis is more analogous to its analog tape predecessor. I.e., It should have made no difference if the power was interrupted at the end of recording, it should still be on the drive and recoverable, at least from the FireWire based adapter? Not sure why you cannot play it back on the original recorder however? You may want to contact Alesis customer support?

Computers are our friends! RIGHT, SURE....YOU BET....
Ms. Remy Ann David

MadMax Mon, 04/17/2006 - 02:12

Remy's right,

The data files are still there.

Let me echo the "be sure you don't take that drive to anothe computer!" thought.

Try this first; do a factory reset... When powering up the unit from the front power button, simultaneously hold the Record and Play buttons. Hold them for about 5 seconds... or until you see the "Restoring Factory Defaults" message.

This might get the recorder back on track to get your tracks.

Get ahold of the Firecard adapter and THEN transfer your files... the ftp process is killer slow! (Unless it's just a couple of tracks, even then, your time is worth something, right!?!)

Good luck!

Max

anonymous Mon, 04/17/2006 - 04:10

My best suggestion is that you join the HD24 group on Yahoo and ask about it there. You'll find others who have been down this road ahead of you. In fact, anyone who owns or uses an HD24 should be a member of that group.

Unfortunately, none of the suggestions above will work to recover your files. The HD24 does require you to press "Stop" to save your files. If the power is interrupted before you press Stop, the data will not be recoverable through any normal means. That's why it's so important to use a UPS with hard disk recorders.

The data is still on your disk, but recovering it will be VERY difficult. It has been done successfully by others, but there is no cookbook approach. It will require you to do some very detailed low level disk analysis and recovery work. There is no simple fix.

Come join the Yahoo group. Search its archives for this topic - there have been a number of discussions of it in the past, some this year, and some in past years. Good luck.

MadMax Mon, 04/17/2006 - 18:25

Gilliland wrote: Unfortunately, none of the suggestions above will work to recover your files. The HD24 does require you to press "Stop" to save your files. If the power is interrupted before you press Stop, the data will not be recoverable through any normal means. That's why it's so important to use a UPS with hard disk recorders.

The data is still on your disk, but recovering it will be VERY difficult. It has been done successfully by others, but there is no cookbook approach. It will require you to do some very detailed low level disk analysis and recovery work. There is no simple fix.

Ooooo, that stunks!~

I was misinformed. I was given to understand that this was one of the slick side-benefits of the Alesis HD24... "even if you loose power, the files are recoverable." My "ASS"umption was that the common reset to defaults would reindex the drive and the files would be there... still got some major learning curve ahead... big shock there, huh?

Thanx for the headsup on the HD24 group! (Just signed up)

... and even MORE thanx for the UPS advisory!!! 8-) I've got a couple hangin' out round here... Gonna go plug em' in in a few minutes.

On the curve,
Max

Cucco Fri, 04/21/2006 - 12:13

Actually, that's what has me puzzled about this post. When I got my HD 24, it was shortly after a product debacle with a Tascam HD recorder. I lost power to it and it threw away all of my data!!!

So, I bought my HD24 from a local dealer who was more than willing to let me return it if I didn't like it. I plugged 'er in, began recording 8 tracks and literally yanked the plug right out of the back of the unit.

Lo and behold, all 8 tracks had recorded and all of the data up until I yanked the plug was securely in tact. It was that moment that began my love affair with my HD24 and the very reason that you'll have to pry it out of my cold dead lifeless hand to ever get mine away from me.....

Oh well, I guess wierd things happen some times...

J.

anonymous Fri, 04/21/2006 - 12:23

Cucco wrote: I plugged 'er in, began recording 8 tracks and literally yanked the plug right out of the back of the unit.

Lo and behold, all 8 tracks had recorded and all of the data up until I yanked the plug was securely in tact.

The HD24 will retain its data once "Stop" has been pressed. It will also retain its data if you are recording over tracks that have already been allocated. So some people advocate preallocating your recording space by making a blank recording of sufficient length before beginning to record actual audio. If you then record into that preallocated space, the HD24 will retain it even if it loses power in the middle of recording.

But if you are making a new recording into unallocated space, you absolutely MUST press "stop" to get the HD24 to update its allocation tables. If the power drops while the unit is still actively recording, the data may still be on the disk, but there will be no way to access it. For all intents and purposes, it is lost.

If you believe that your test gave different results from this, then I strongly recommend that you go back and rerun your test.

anonymous Sat, 04/22/2006 - 13:10

MadMax wrote: Could this be a firmware thing?

I doubt it. The Alesis team that develops the HD24 are regular participants in our HD24 group on Yahoo, and they've never mentioned any such firmware changes. And considering the number of people who have encountered this failure, I'm pretty sure they'd have brought it up if they'd figured out a way to solve it. Alesis has been pretty good about making firmware improvements and bug fixes, but this is one issue that can't be readily fixed in software.

Davedog Sat, 04/22/2006 - 14:27

This is interesting to hear, since I too have an HD24. A similar situation happened to me in that a power loss occured(I kicked out the plug under the desk) while recording some test tracks. The tracks were unlocked at 24 and nothing other than the redlight was on. Analog input . The tracks came back up at restart and were no good to me, but I could play them back,monitor was as regular,I could have tracked to them, editted them. They were only the length(of course) that the power was on....there were NO preliminary tracks though I DO name my tracks before I begin recording but establish no particular length for them. Perhaps the NAMING does the trick....Interesting thread.

MadMax Sat, 04/22/2006 - 20:02

I've gotta think that there's something about it, other than the naming...

I'm gonna try a couple of permutations in settings tomorrow (Sun) afternoon or evening. (Got some honey-do to take care of during daylight...)

1. Digital input; 24 tracks@48, No Name (As it comes out of the box)
2. Digital input; 24 tracks@48, Named Song
3. Analog input; 24 tracks@48, No Name
4. Analog input; 24 tracks@48, Named Song
5. etc.

After 2 minutes, I'll pull power. (I've got two units and they're both still under warranty)

It might be Monday before I can post the results, but I'll post what I get.

Max

anonymous Sun, 04/23/2006 - 06:28

Davedog wrote: This is interesting to hear, since I too have an HD24. A similar situation happened to me in that a power loss occured(I kicked out the plug under the desk) while recording some test tracks. The tracks were unlocked at 24 and nothing other than the redlight was on. Analog input . The tracks came back up at restart and were no good to me, but I could play them back,monitor was as regular,I could have tracked to them, editted them. They were only the length(of course) that the power was on....there were NO preliminary tracks though I DO name my tracks before I begin recording but establish no particular length for them. Perhaps the NAMING does the trick....Interesting thread.

It IS interesting. There have been so many people who have lost tracks this way that it would be VERY interesting to learn if there is a technique that helps the HD24 avoid this problem.

Whatever testing you do, please post your results to the HD24 group. Let's get the benefit of the entire HD24 community to discuss and analyze this. Here's the group:

http://groups.yahoo.com/group/hd24/

We've got over 1200 members, so it's a pretty lively group of HD24 owners. Might as well take advantage of that much knowledge and brain power!

anonymous Sun, 04/23/2006 - 14:58

yahoo group was a life saver...

thanks for your tips fellas, i ended up calling alesis tech support who pointed me to the yahoo group. i thought it was funny that the guy on the phone hadn't visited the site enough to tell me, but there ya go.

anyways, after trolling the group for a bit i found the solution involving block copies, and was able to recover the entire performance! thanks for your input and direction, i definitely recommend the group, there are a lot of very knowledgeable users who have come up against any number of problems. i recommend searching there for a fix before bothering with tech support.

anonymous Sun, 04/23/2006 - 15:54

Re: yahoo group was a life saver...

esquire97 wrote: anyways, after trolling the group for a bit i found the solution involving block copies, and was able to recover the entire performance! thanks for your input and direction, i definitely recommend the group, there are a lot of very knowledgeable users who have come up against any number of problems. i recommend searching there for a fix before bothering with tech support.

Excellent! I'm so glad to hear that you got your recording back. Not everyone has been so lucky. That's great news!

I'd recommend that you document the steps that you took and post them back to the group - it will help the next person who has a power failure.

MadMax Mon, 04/24/2006 - 18:35

As promised, I'm posting the results of my "Alesis HD24 Power-Loss" Test.

This test is the result of a posting to the "Pro Audio Gear" forum at [[url=http://[/URL]="http://www.recordin…"]Recording.Org[/]="http://www.recordin…"]Recording.Org[/] in which a member had experienced a power failure during a recording session and was at that time unable to recover the audio files.

The member was eventually able to recover said files, but not before a series of comments was made whereby another forum member (and RO Moderator) had stated that he was able to recover files after a power failure.

As a relatively new HD24 owner, I was under the impression that the "automatic" file recovery was a normal situation with the HD24 family of recorders.

First let me state that I am hopeful that the results will spawn a couple of other HD24 users to either take hede of the results, and/or continue to explore why some users of the HD24 system get different results.

While each test in the permutation was different, the result was the same. TOTAL data loss as detailed!

The test setup and specifications:

OS Version 1.20
Boot Version 1.00

HDD: Factory Installed Unit
Maxtor 2F040L0 (41.1 Gb)
VAM51JJ0
File System 1.10

Clock - Internal
Monitor - Auto

Test audio was captured for a period of time between 30 and 60 seconds. The same audio source was used for all tests. At the end of each time frame, power was summarily pulled at random, to simulate a typical power outage... (and because it took a random time to walk around to the back of the unit.)

After each power outage, the power switch on the unit was turned off and then power was restored as normal. On power up, the Factory Default reset was pereformed.

Phase 1 - Various configurations of digital input. Tracking 8, 16 and 24 channels of ADAT lightpipe audio. As I do not currently have access to a 96KHz lightpipe source, I was unable to substantiate a valid test set at 88KHz or 96KHz.

In Phases 1 and 2, A new song was created in the usual manner, without naming the song, as had been proposed as possible variable.

Each test, I attempted recovery of the tracks using the Onboard Scan & Repair Utility. The utility failed 100% of the time to yield a recovered set of tracks.

After each test, the HD24 was regularly power cycled with a Factory Default reset and the "corrupted" song was deleted and a new song created according to the table(s) below.

Test 1 - 24 tracks@48KHz
Test 2 - 24 tracks@44KHz
Test 3 - 16 tracks@48KHz
Test 4 - 16 tracks@44KHz
Test 5 - 8 tracks@48KHz
Test 6 - 8 tracks@44KHz

Phase 2 - Various configurations of analoge input. Tracking 8, 16 and 24 channels of analoge audio.

Test 7 - 24 tracks@48KHz
Test 8 - 24 tracks@44KHz
Test 9 - 16 tracks@48KHz
Test 10 - 16 tracks@44KHz
Test 11 - 8 tracks@48KHz
Test 12 - 8 tracks@44KHz

Phase 3 - An exact duplication of Phase 1, except that each time a "New Song" was created, the song was named.

Test 13 - 24 tracks@48KHz
Test 14 - 24 tracks@44KHz
Test 15 - 16 tracks@48KHz
Test 16 - 16 tracks@44KHz
Test 17 - 8 tracks@48KHz
Test 18 - 8 tracks@44KHz

Phase 4 - An exact duplication of Phase 2, except that each time a "New Song" was created, the song was named.

Test 19 - 24 tracks@48KHz
Test 20 - 24 tracks@44KHz
Test 21 - 16 tracks@48KHz
Test 22 - 16 tracks@44KHz
Test 23 - 8 tracks@48KHz
Test 24 - 8 tracks@44KHz

Phase 5 - This set of tests, while abbreviated, yielded the same net results as all the others, and had little time left to complete the permutations.

This phase I believe more closely resembles real world scenereos where levels are checked for playback before an actual recording session.

After creating a new song, 60 seconds of audio was captured. Afterwhich, the song was rolled back to "Locate 0". Then the test proceeded as previously outlined in Phases 1 through 4, with the exception that the second recording attempt, was made with a "second" existing set of 8 tracks. That way, differentation could be established as to what was actually captured.

Test 25 - 8 Tracks@48KHz, Digital Input, No Name
Test 26 - 8 Tracks@48KHz, Digital Input, Named
Test 27 - 8 Tracks@48KHz, Analog Input, No Name
Test 28 - 8 Tracks@48KHz, Analog Input, Named

I should note, that each time the unit was powered back on, the "song" did have the correct corresponding amout of time initally allocated to the song. When played, the song was playable up to the initial 60 second timeframe, or shorter, when the power outage was performed.

In each case, the audio captured on the initial timeframe was the audio which was played back, and NOT the second recording attempt. This was true before an after each attempt at rovery with the Onboard Disk Utility.

This phase, while incompleted in term of running the permutations, does show promise that there may be a set of criteria that may yield recoverable audio tracks. But at this point, it is purely speculation, and down-right hopefullness on my part.

If you are an Alesis HD24 owner, and have differing specifications of OS, or another HDD type/size, I would encourage you to continue this set of permutaions and post it to this thread.

The continuation that I would encourage would include differing HDD's, 88 and 96KHz sample rates and external clock.

Thanx for your attention through all the boring details,

Max
(DM Mobile Recording)

RemyRAD Mon, 04/24/2006 - 20:37

It sounds to me that one of the factors involved here may be that of the disks already having been formatted or not formatted?

I can only imagine that if one is recording on a hard disk drive that had been recorded on before, then it would have been formatted successfully. While recording data upon an already formatted drive and then having the power interrupted it may not necessarily delete or corrupt the disk format, creating recoverable data? But if you are recording on a virgin disk drive and are formatting while recording while power is interrupted, I can understand why tracks may not be recoverable on a disk that has not completed its format utility function.

Nobody mentioned this so I thought I would just cause trouble again?
Ms. Remy Ann David

MadMax Mon, 04/24/2006 - 20:48

Remy,

The very first test in Phase 1 WAS a virgin drive!

After that, it was... well... I won't say I was too kind when I violated it, but I was only as brutal as I had to be.

I bought two drives with my first unit. I hadn't fired the second one up yet. So essentailly, I wanted the very first test to be as close to out of the box as I could get... Fire the HD up, format it, create a new song, put it in the default 24 track@48KHz digital input mode and go!

Max the (drive) Masher

RemyRAD Mon, 04/24/2006 - 21:30

But Max, now I'm Mad! You did say you formatted the drive before these tests. Was that what was done before? Your tests seem to be very thorough but this formatting thing is a factor I believe? I just really want a pair of those machines!

Never enough for me. Size does matter!
Ms. Remy Ann David

(Now people will think I'm cheap...?)

MadMax Mon, 04/24/2006 - 21:49

I'm rather high maintenance myself... Just ask the Mrs!

Seriously though, you MUST format the drive before it can be used.

If there's a way to just slap a drive in and take off on tracking, I didn't see it... but I've been known to overlook some blatently obvious stuff before...

I'm really happy with the two HD24's I've got. The only thing I wish they had was a built-in SMPTE sync. Then I could get into some POST and A for V stuff w/o too much hastle. Tracking a SMPTE code to a track, like the "good ol daze" of ADAT's... just seems like a step backwards... again.

May have to look into hetting a couple of MX2424's... IF I can find em'. That's another thread, huh?

Slappin' my high maintenance self silly,
Max

anonymous Mon, 04/24/2006 - 22:03

RemyRAD wrote: But Max, now I'm Mad! You did say you formatted the drive before these tests. Was that what was done before?

No, that's not an issue. The HD24 requires you to format the drive before recording on it. And these failures have been reported on drives that are freshly formatted as well as those that have had data written to them and then erased. So the relative "virginity" of the drive doesn't appear to be an issue.

I'd like to hear more from Cucco about exactly what might be different in his scenario. I don't want to disregard what he's telling us, but there are SO many people with the opposite experience that it's hard to take his report seriously. Still, I really hope he'll come back with a more detailed explanation. It would be great to learn of a way around this problem - scores of HD24 users have been burned by it.

RemyRAD Mon, 04/24/2006 - 22:07

I know it requests that you format the drive first, but that is just for the "header" or perhaps the table of contents, since it only takes moments? Because of the way the Alesis works, it is only fully formatted after the disk has been initially fully recorded upon and not just the header, perhaps? That is may be why everything is occurring the way it has? I think only Alesis could verify this?

I don't care I still want a $1500, 24 track machine!
Ms. Remy Ann David

anonymous Tue, 04/25/2006 - 05:43

RemyRAD wrote: I know it requests that you format the drive first, but that is just for the "header" or perhaps the table of contents, since it only takes moments?

The Alesis allows you to do either a long format or a quick format. Most of us typically do the long format when we install a new drive, just to make sure that the whole drive is properly initialized. Still, all that really does is verify that all the sectors can be written successfully. IDE drives come from the factory with their sectors preinitialized, so it really doesn't change anything. For the purpose of this discussion, it really doesn't matter whether you use the long or the quick format.

Davedog Tue, 04/25/2006 - 16:25

So ,Max, as I understand it, there was never a time in your test that the audio was still there???Except when you allocated a length of time for audio, recorded it, went back to zero and began again, kicking out the power which in turn did NOT allow the second set of audio but retained the first set......

If this is the case then Gilliland's point of needing to push STOP must be true.

My accounting of my experience goes no further than the one time and seems unexplainable. I had no desire to see if it would happen again and as it was at install its not possible for this to happen in the future. I have a small UPS on the recorder...have had for ALL the recorders over the years...invaluable tool BTW.....And its unlikely we would lose power as we are in a fairly stable and modern grid here at the ranch.

I need to visit the HD site as I do have some small concern over my HD freezing just after start-up and then after a reset NEVER freezing again till the next power-up. Its not a problem as it is lossless in its nature ...just annoying and as I expect it , its simply a matter of waiting out its little game and then proceeding with the session.

Remy.....You will like this machine in that its much like a tape machine.... its a destructive recording device. There are no virtual tracks and hidden stuff...Its 'do it right or do it over'...much as it SHOULD be anyway.....That is if you want great recordings.

MadMax Tue, 04/25/2006 - 17:28

Davedog wrote: So ,Max, as I understand it, there was never a time in your test that the audio was still there???Except when you allocated a length of time for audio, recorded it, went back to zero and began again, kicking out the power which in turn did NOT allow the second set of audio but retained the first set......

If this is the case then Gilliland's point of needing to push STOP must be true.

Unfortunately, as far as the tests went, you are correct... but alas, I wish I weren't.

Still, there's GOT to be something in both yours AND Jeremy's accounts!!! It couldn't just be a fluke of nature... maybe the Recording God's smiling favorably, but not just an annomolly(sp?).

There's something about both of your units that we just don't know about, or something about the exact way that preliminary events occurred... but the J-Man can duplicate it... Hopefully he'll be back soon and can chime in.

I really believe that there's SOMETHING to this... I KNOW you guys have nothing to gain from BStin' us. So, it's not a matter of integrity. It's a matter of digging out the details.

My accounting of my experience goes no further than the one time and seems unexplainable. I had no desire to see if it would happen again and as it was at install its not possible for this to happen in the future. I have a small UPS on the recorder...have had for ALL the recorders over the years...invaluable tool BTW.....And its unlikely we would lose power as we are in a fairly stable and modern grid here at the ranch.

Yeah, I work w/o enough nets, waaaaaaay too often. I put a UPS in the rig. I'm currently not tracking here in the house, just mixin', so if I loose power... hardly ever.

I need to visit the HD site as I do have some small concern over my HD freezing just after start-up and then after a reset NEVER freezing again till the next power-up. Its not a problem as it is lossless in its nature ...just annoying and as I expect it , its simply a matter of waiting out its little game and then proceeding with the session.

You really should! LOTS of good info... even about being able to remote control... W/O that BeastRC!!!

Remy.....You will like this machine in that its much like a tape machine.... its a destructive recording device. There are no virtual tracks and hidden stuff...Its 'do it right or do it over'...much as it SHOULD be anyway.....That is if you want great recordings.

Amen, bro! It ain't a RADAR, but it don't cost what-a-radar-do, either!
Darn thing's not soundin' too bad at all. I'm dumpin' a coupla' projects from ADAT's in and the gain is very suprising!

Doing locate's and loops is a be-ach w/o a remote, but the BeastRC software in mine is just toooooo unstable.

Remy, you GOTTA snag a coupla' these puppies!

Max

MadMax Tue, 04/25/2006 - 17:37

Gilliland wrote: The Alesis allows you to do either a long format or a quick format. Most of us typically do the long format when we install a new drive, just to make sure that the whole drive is properly initialized. Still, all that really does is verify that all the sectors can be written successfully. IDE drives come from the factory with their sectors preinitialized, so it really doesn't change anything. For the purpose of this discussion, it really doesn't matter whether you use the long or the quick format.

OOoooooooh, possibly the start of another test here...???

I only did the quick format... again, trying to get as reasonably close to "Out of the Box" virgin as I could.

I've got gigs all weekend and mixing most of next week, but somewhere in the next week or so, I'll snag another new drive and do a full-on long format and see what happens... I'd like the permutations to be as complete as possible...

Not quite OCD... but DEFINITELY anal,
Max

anonymous Tue, 04/25/2006 - 20:10

MadMax wrote: [quote=Davedog]Remy.....You will like this machine in that its much like a tape machine.... its a destructive recording device. There are no virtual tracks and hidden stuff...Its 'do it right or do it over'...much as it SHOULD be anyway.....That is if you want great recordings.

Amen, bro! It ain't a RADAR, but it don't cost what-a-radar-do, either!
Darn thing's not soundin' too bad at all.

Remy, you GOTTA snag a coupla' these puppies!

Max

And let me reiterate that anyone who owns one of these needs to join up with the Yahoo HD24 group. We've got a huge community (over 1200) of HD24 owners there, so the knowledge base is considerable.

One other thought - anyone who is buying an HD24 these days should be sure to get the XR version. It doesn't cost that much more, and the better converters in the XR version make a considerable difference even at the 44.1 or 48k sampling rates.

anonymous Tue, 04/25/2006 - 23:09

Hi, my first post here.

Interesting discussion. Could the HD's write caching policy a factor here?

In windows you can enable/disable HD write caching, so power failure won't result in data loss.

Go to "Device Manager -> Disk Drives -> Drive -> Policies" and see some explanation there.
I wonder if HD's with different cache sizes/write caching rules would dictate at what point your data is actually written to disk.

-teflon

anonymous Wed, 04/26/2006 - 02:30

teflon wrote: Interesting discussion. Could the HD's write caching policy a factor here?

Interesting question. To the best of my knowledge, hard disks don't do any write caching. They use their cache to improve read performance only. Many operating systems do write caching, usually with an option that allows an application to disable it when necessary. But I've never heard of a disk that would cache its writes independently. Does such a thing exist?

MadMax Wed, 04/26/2006 - 04:36

Gilliland wrote: [quote=teflon]Interesting discussion. Could the HD's write caching policy a factor here?

Interesting question. To the best of my knowledge, hard disks don't do any write caching. They use their cache to improve read performance only. Many operating systems do write caching, usually with an option that allows an application to disable it when necessary. But I've never heard of a disk that would cache its writes independently. Does such a thing exist?

AFAIK, ALL drives with ANY amount of independant processing, have some amount of R/W caching... albiet very little in most cases. (Actually I think anything after RLL's have W/R caching) Although, I don't like the fact that the industry has called it caching. It's more of a buffer. If it's data, it goes in there. Again, AFAIK, I don't think command/control goes through the cache'.

All of the SCSI drives I've ever seen have a minimum of 16kb... all the way up to 64Mb!

I think - I'll check later today if I get a chance - most SATA's run something like either 16, 32 or 64 megs... thus the performance in an EIDE bus architecture.

Some of the bigger/better made drive$ have a $mall battery backup to hold duplicate/backup personality info. It's also claimed to be capable of holding the cache until the end of xxx W/R's. It's supposed to allow/force an ALL Write command so save it to the drive... again, it's a small buffer, so it's supposedly able to write it to the drive in just a few cycles. Some more of that HDD voo-doo that they doo-doo.

I'm convinced that some of these folks that design and write hdd controllers, drives and I/O's just ain't right in the head! I hear they just slide pizza under the door to feed em' and don't allow them in public unless it's by accident... and then... ONLY at night!

Max

anonymous Wed, 04/26/2006 - 20:17

EDIT: I posted a question about this on the PC-DAW mailing list, and was told basically that I am wrong about the issue of IDE Write Caching. Apparently IDE drives DO perform some degree of write caching. So I stand corrected. The fact then that IDE drives DO have differing write behavior certainly does add a new wrinkle to our discussion of how the HD24 copes with power failure scenarios.

I've deleted the prior content of this note (originally posted about six hours ago) as it was simply incorrect.

MadMax Thu, 04/27/2006 - 04:31

I too need to stand somewhat corrected on the caching issue...

While all drives have SOME amount of caching, I cannot find, so far, any indication of SATA/SATA II, drive caching amount.

I've heard, as I posted, anywhere from 16-64Mb, with the general understanding that the command and control does NOT go through the cache.

Evidently, the command and control are put through the buffer as well as all the data! And the SATA standards committee now refers to it as a buffer!! However, as I write this, I have not seen any references to any amount of buffer that is used... so it could be anywhere from 8k to ??? As fast as the W/R process goes, I don't realitically see where it could be any smaller than 8Mb... The SATA drive transfer rate is greater than 60Mb/sec, (1Gb/Sec), so they couldn't put enough data through without much less than 32Mb, but that is only speculation. (SATA II is an astounding 3Gb/Sec)

I know all this is not necessarily related to our thread, but I had made an assumptive statement, and wanted to clarify the probable inaccuracies.

Max

anonymous Thu, 04/27/2006 - 11:09

Hello, I'm new to this group. I couldn't help noticing this thread.

There seems to be some cross-posting between this forum and the yahoo groups, so Esquire97, if you happen to be Laurent I'll let you know when the disks arrive.

If you have no idea what I'm talking about, then you must be someone else and I may be able to help out recovering the lost data.

Just give me a shout.

Marc

x

User login