Sata Problems!!!

[ChrisBlackwell]

Howdy, here is my sucky problem:

First off, my specs:
ASUS P4C800 Motherboard (I'm sure there's a zillion different p4c800 models but i can't find any other numbers right now)
P4 3.2GHz 800MHz FSB
MSI 256MB GPU (Again, I'm tired of wrecking my knees looking for the number)
2 x 17" Samsung LCD's
2 x 120 GB Seagate SATA Drives
Various other unrelated peices of hardware (DVD, Firewire/USB combo, key/mouse combo, blah, blah)
350W PSU

Also, I produce music with my computer with:
MOTU 896HD (Firewire High-Definition Audio Interface)
MOTU Midi TimePeice AV (Parallel Port Midi Interface/Syncronizer)

So here's my problem...

For probably 3 years I was using my harddrives on an integrated RAID controller on my motherboard... I had it set up for stripe, and a very delightful setup indeed... worked great, no problems...

Life got in the way and I had to leave my computer with some butcher (ie a friend who thinks he knows alot about computers, but actually he know alot about wrecking computers)... when i went back to retrieve it, the afforementioned butcher told me that it had just stopped working... i took a look at it and sure enough it was toast.... the asus picture thingy came up, then the RAID utility thing said "no drives" or something, then stopped...

i restarted, same thing
i restarted again, went into setup, and again, setup said no drives...

so since it was no working anyway, i set the raid controller to just normal ide mode....

still nothing...

now, this motherboard of mine also has non-RAID SATA plugs, so i switched over to there to see what happened... (note: i turned the raid controller off in the setup)

lo and behold there are my drives.... hmm... interesting...

out of curiosity, i went back to the RAID plugs, turned the raid back on in setup, and voila, the whole thing worked again...

booted up into xp, everything looked fine....

use it for a few minutes, and the harddrives freeze again (mouse pointer still functional, but as soon as you click or do something that the OS needs to go to the drive for, it kaks)

bummer.... so i figured the harddrives were toast...

when i got back to my home town, a dude i know at a computer store said that seagate drives have a five year warranty, so i gave them to him, and he sent them out to be refurbished... many weeks later they just gave me new ones so i didn't have to wait anymore (they were definately new, the bags were sealed up, etc.)

so i put them in, installed xp, everythings ok, then one day... kaput

so i tried the same thing, turned off the raid and plugged them into the non-RAID SATA plugs, and they were detected... so i went back to the RAID and everything was ok... for a while and then kaput

so, now i'm thinking that it's the raid thing... so i did all i could to get all my important stuff of the harddrive and put it on my ipod, switched to the non-RAID SATA (which i was not happy to have to do), installed xp, and same thing... after a while it just kaks... and when i restart, they don't detect...

so this is starting to annoy me! i figure it's not the drives, because a) what are the chances that two drives fail at the same time, and b) what are the chances that two drives fail at the same time TWICE and c) in my experience, when a drive fails, that's it, it's done, it ain't coming back... and mine always come back...

so what i've got into the habit of having to do is whenever this happens, first i shut off the tower, then i flick the little switch below the power chord off and then back on... sometimes this works... other times i have to open the case, pull the SATA plugs out and plug them back in... sometimes this works... somethimes i have to leave the computer off for a couple hours and then it'll work again...

either way, this situation is completely unacceptable... i do recording for money with this machine, and i have also done live sound with this machine.... this cannot happen to me in front of 500 drunken punkers!!!

i've read a few posts in this forum and i'm beginning to think my power supply unit is inadequate, but it was ok for like 3 years, and i really doubt that's the problem...

please help me! i shudder to think that i have to rely on my laptop to do my recordings (2.8GHz don't cut it!!!) wtf is wrong?

[Xpenetrator]

Your "butcher" might have fried the mainboard (static), but this sounds unlikely when your RAID and non-RAID controllers are actually different chips and not combined in the same one (don't know). A good thing would be to interrogate him what exactly he tried to do with your computer and for what reason.

Shot in the blue:
I have no experience with S-ATA troubles yet and I don't know if S-ATA cables with extreme low quality like some IDE actually exist - but have you tried replacing one with a cable fresh from the store? They are a "common denominator", too. Some IDE cables didn't survive being touched more than once and bending them wrong could produce very similar effects like you described. Would be nice if it's that easy...

Various other unrelated peices of hardware (DVD, Firewire/USB combo, key/mouse combo, blah, blah)
350W PSU

You have quite a bunch of power-hungry hardware (depending on your GPU, too) that only a 350W PSU of best quality would feed sufficiently. Did you know that PSUs can get weaker/change their properties over the time? Maybe the load on the 12V during startup of all the motors etc. prevents something from being initialized correctly all the time?

[ChrisBlackwell]

Thanx for the reply...

No doubt the butcher was sufficiently interrogated... trust me, if he was smart enough to know what he did wrong, he wouldn't be dumb enough to tell me...

yeah i thought about the cable thing, too... unfortunately, these problems have left me unable to make money, and i don't even have the funds to buy new cables and find out... however, assuming the unlikelyhood that both my cables are bunk, i've been just using one harddrive and switching the cables around to see if one is better than the other... i thought i had actually solved it but just now it kacked again... either way, it's been my experience that if it works at all than it probably works (i keep saying that but ovbiously something is working for a while and then kacking :( )

anyway, thanks for replying

[Xpenetrator]

No doubt the butcher was sufficiently interrogated... trust me, if he was smart enough to know what he did wrong, he wouldn't be dumb enough to tell me...

That's a pity because every little detail on what he did could help. He could have tried to tune some settings wrong in the BIOS ("I'm gonna make your computer faster") or did he even open the case so that he could have damaged something physically?

however, assuming the unlikelyhood that both my cables are bunk,

That assumption once costed me a lot :) If both cables were touched the same wrong way they can be equally broken..:mrgreen: But I doubt that, too since they are much less sensible by design. I just thought that was an easy thing to try.

I found very few problems with P4C800 RAID controllers on Google (unlike the A8N SLI-Deluxe, there's much to read about problems, even similar to yours), this seems to be no common problem. No big help. But this and the fact that both independent controllers show the same thing in RAID and non-RAID, makes me think that the mobo could be defective.

Have you contacted Asus Support already? They know how to fix that maybe, if it could be fixed.

For the case that something was misconfigured by that guy, have you tried resetting the BIOS and the controllers to its defaults already?

Does the Event Viewer report something when the drives stop working during operation? (Go Start->Run, type eventvwr.msc <Enter>, in left tree view select "System" and browse through the red X marked 'error' entries, if any. If one matches the time your HDDs bailed out, doubleclick to unfold the message.)

[ChrisBlackwell]

no, the bios is set the way it always has been, i made sure of that... no overclocking, no other weird settings... that was one of the first things i did actually was reset the bios...

the thing is, i've freaked out on that guy a number of times before he got a hold of that computer... i don't think he tried to change anything or open the case up because he knows i wouldn't be happy... i believe him when he says he didn't do anything out of the ordinary... when i say he's a butcher, it's because he has this uncanny ability to wreck computers just by looking at them (i'm sure we've all met at least one of these people)... he's one of these guys that will delete directories from his program files directory because "i don't have time for add/remove programs", bullshit like that... but, as far as he told me, when it started acting up, he just left it alone because he knew that no matter what he did, i would probably blame him for wrecking it (i've made him a bit paranoid, thank god)

i was talking to a guy at the local computer store today and he said that in his experience there are alot of problems with SATA cables, so that being the cheapest thing to try, i think i'll go with that first... the other thing he said, like you did, was about the psu degrading over time...

thanks tho, i'll let you know what happens when i get new cables (fingers crossed)

[Xpenetrator]

I'll cross my fingers for you, too :mrgreen: (You got my full compassion, I'm in the same kind of business...)

[ChrisBlackwell]

i haven't bought the new cables yet, but there is definately something going on with one of my cables... when i use the one cable i hear clicky noise coming from the harddrive, and not when i use the other.... however, the same thing happens in windows no matter what cable i use

[Xpenetrator]

If nobody touched them before the problem came up, the cables get a bit out of the focus - they normally don't fall apart without help. Another idea would be to test the drives _plus_ the cables (second choice: the cables alone) on another computer, to rule them out.

[ChrisBlackwell]

all my friends are poor, they've all got p2's :s i don't know anybody else with SATA... i probably would have figured this out a long time ago if i knew somebody with SATA drives... oh well... i still doubt it's the cables, honestly... i figure that's a pretty lame thing to happen, considering all the complex things that can go wrong with a computer, but like i said, it's the cheapest thing for me to check right now... i have a friend that works at the board of education up here, he says he's gonna score me a better psu and see if that makes a difference... i was thinking of blowing out the one i have in there now... i've had to do that to the processor fan a couple times...

[Xpenetrator]

I'm still on P-ATA / IDE, too...:) It serves me 50+ audiotracks which should be roughly enough for simple productions ;) Blowing away the dust is always a good idea. Dust plus humidity (or worse: plus nicotine) deteriorates the cooling performance and can make a fine, slightly conducting smear that can be really bad if piling up on the wrong place of a PCB. I'll cross my fingers again that the PSU would help.

[ChrisBlackwell]

i don't smoke (cigarettes, hehe :P)

[Chaky]

I smoke (cigarettes too).

I vote for PSU replacment. Wouldn't cost you more than $30 to get the 550W like I did.

It is really frustrating when you must guess the power consumption of your hardware. I'm wondering, is it really that hard to declare power consumption of a HD or GPU, or mobo... Would make things simpler when it comes to building PC's.

[Xpenetrator]

I smoke hair spray...:mrgreen:

It is really frustrating when you must guess the power consumption of your hardware. I'm wondering, is it really that hard to declare power consumption of a HD or GPU, or mobo... Would make things simpler when it comes to building PC's.

Unfortunately I haven't seen something like this http://www.meisterkuehler.de/cms/energieverbrauch.html in english yet. (Scroll down the page a bit to see the power consumption calculator. You'll get some typical numbers for the various components this way, too)

[ChrisBlackwell]

that's some scary shit

[Chaky]

Thx for the link.

Too bad that it is in german and outdated. (didn't find Athlon 64 +3700 nor GeForce 7800 GT, closest one was High-end vga with 512 MB)

AMD athlon +3700
PCTV SAT
1 GB network card (I've selected 2 for that reason)
1 firewire
1 sound card
1 sata drive
1 IDE drive
1 vga with 512 MB
1 DVD burner
1 CD rom
1 Floppy
all kinds of fans
10 (I think) USB ports
Modem

Turns out 352 W

More than one in question here.

[Xpenetrator]

More than one in question here.

Nope. The only one is they assume 140W for the "High end-512MB" - the 7800GT 512 MB takes only 55W according to this article.

I'm pretty sure your 3700+ does not consume more than 90W. The whole calculation has some margin of error and will give you a good approximation only. The result is a "worst case" maximum value.

[Chaky]

My is GAINWARD GeForce 7800 GT 512 MB and it supports dual DVI unlike one in the test.
Nevertheless, I would be very pleased if it is 55W. That would mean that SLI would not require mo' power.

Taking in consideration possibility of OCing, I am convinced that the calculation is far from exact. I think that it should be used for approx. projection of power consumption when building PC.

It gives you a picture of what the PSU should and shouldn't be. I like that.

[Xpenetrator]

I think that it should be used for approx. projection of power consumption when building PC. It gives you a picture of what the PSU should and shouldn't be.

That's what the introductory article on top of that page says. For that purpose, +/- 20W doesn't matter. The other purpose of the page is to calculate the annual costs for the juice. I'm not sure if I really wanted to know that...:eek:

Yes, the G7x chips are pretty power-saving and cool running GPUs (latest 90 nm technology), but I wonder a bit about that 55W, too. But AFAIR they need less or at least not more power than the fastest Geforce 6 cards.

[rsumner]

While I don't have an answer to your problem I thought I would describe my recent experience and what I have found so far.

A couple of weeks ago my Win2K system went down hard. Missing OS files during boot etc. In other words it was toast. After re-installing the OS I discovered that a 1GB memory module was dead. I also noticed that one of the three IDE hard drives in the system kept seeking randomly as if it was struggling, so I decided to get another one of those too.

Since I didn't want to keep buying the same old IDE hard drives I picked up an SATA drive (Seagate 750GB) and an Adaptec DuoConnect SATA board that plugs into a PCI slot. I quickly found that I was unable to format the new hard drive and that it was seeking randomly also. Using another PC I powered the new drive by itself and it stopped its seeking problem. So then I picked up a new power supply and no more seeking problem.

Unfortunately, I introduced a problem unknown to me by my upgrade. One that I have come to believe is somewhat common with these new SATA drives and controllers. My audio card (Soundblaster Audigy 4) clicks, stutters, hesitates and pops a lot. Especially if I do anything that causes the system to work harder while its playing back a sound file, open another program etc. This is a real bummer as I used this system to record a lot of old record albums. I never had any problems with the sound before installing the SATA hardware, even with the other power supply which does appear to be bad. Researching this on the internet I have found several posts from people who describe the same problems or try to explain why the SATA hardware is causing this. Basically, the only real solution I have heard so far is to get a motherboard that has a built-in SATA controller "On its own bus" and not sharing the PCI bus with your audio card.

[Xpenetrator]

Thanks for your report. This confirms the suspicion that the threadstarter's PSU is the problem.

Your sound problem can maybe solved by changing slots. I don't know your mobo, but the scheme is like this:
The bus has 4 hardware INT (errupt) lines (not to confuse with IRQs), called INT A to INT D. When two busy devices share one INT line to assign IRQ requests, the problem occurs. Your soundcard is sharing an INT line, probably with your S-ATA PCI card.
Refer to the mobo manual to find a PCi slot that shares with a device you don't use or that's less bandwith consuming, maybe with a WiFi device or the LAN. So the problems you were finding in the net were not necessarily due to the S-ATA in particular, but with this INT-thing in general.

There is one more thing that could render the theory above useless: I'm not shure if Audigy cards are as avid on PCI resources as their precessors. That means that a bandwith-hungry S-ATA controller plus the Audigy can have problems, regardless if they share an INT line or not. But try the slot shuffle and maybe it helps. Good luck!

[Chaky]

I had similar experience with PCI video card on my old machine (ATI Rage IIc). The problem is that PCI card that is bandwidth-hungry (VGA or SATA controller) tends to lock the PCI bus while it is working to get that 0.001% better performance. That causes all other PCI based equipment (sound card included) to "pause" for that split-second.
Only remedy suggested was to find and change the setting for that "exclusive PCI" something-mode, or to upgrade (in my case VGA) firmware and/or drivers that would enable you to change that setting.