Corrupt Raptor? Help!

[Blown_65]

Bought a Raptor 74G from Newegg... I couldn't resist the hype... I installed XP on it not once, not twice, but three times and it just doesn't reboot after I begin installing all my programs!? (a routine I've done MANY times w/out problems) Installation of XP was faster, not as fast as others made it seem... The little blue bar (when loading windows) goes by 4-5 times... while my sataII drive was 3.5... (others post it only passes once)->Which mine did the first few boots... Then:

Windows could not start because the following file is missing or corrupt:
\Windows\System32\config\system

It did this each time I installed different programs so it wasn't any certain one... plus after the 3rd install of XP some programs wouldn't even load!?
ANYONE?? HELP!! Should I RMA it? Refund? Will a good raptor be worth it? I ran it through a check/scan disk and it came up fine? SMART shows spin up time value as 2300 or something!? HHEEEEEEEELLLPPPPP!!!!

 

[woshitudou]

Try replacing your cable, it could be probs with that or maybe check what jumper options you have. If you have a sata2 drive on a sata1 mb you have to jumper the drive to sata1.

 

[Blown_65]

It's on a Gigabyte 965P-S3 (w/sataII)... I tried two different cables... not luck. I wouldn't think I'd need to mess with jumpers in my situation. It's passing all WD diagnostic checks... is there any way possible it's my Windows installation disk? (I've never had probs with it before though...)
?????????????????????????????? $150 down the drain!!!

 

[woshitudou]

I've experienced this problem before but I just can't recall what I did to solve it. It may have been the RAM messing up and ruining the system or it could have been my power supply being too weak and fluctiating the power to the hdd.

Maybe try unplugging your dvd drive and any extra drives and having only your hdd powered on whichever power cable coming out of your powersupply. Also try another lead from your powersupply just incase one of your v rails is over burdened.

 

[Blown_65]

I'll have to give that a try... Do Raptors need more power than regular drives? I've had two drives in there before w/no probs either... I'm running a Thermaltake 430W which is powering.... Conroe6300(stock clock)/SonyDVD-RW/WD250WKS/WD74Raptor/5 Fans/7600GT... My ram is Geil DDR2-800(stock clocks)... Thanks for the replies! I guess I have more troubleshooting ahead...

Oh, and I'm not sure how loud Raptors are supposed to be, but sometimes during benchmarking/testing it gets crazy loud! My other WD drive I can barely hear.

 

[woshitudou]

A drive which spins faster takes more power than a slower spinning drive but not so much of a difference. Hopefully unplugging some extra stuff will be the answer, it looks like your 430w could be under strain from all that hardware.

 

[cs120ban]

It is possible the MBR is messed up. You need to zero out the HD or first 30% where the MBR is located and then reinitiate. I use gdisk.exe.

 

[mkaibear]

Download Western Digital's hard drive testing utilities from their website and test the drive.

 

[evongugg]

I agree with mkaibear about testing the hard drive with the manufacturer's diagnostics.
Also I recommend doing a smart test, with the Hitachi Drive Fitness Test.
You can google it.

 

[Blown_65]

Thanks for all the replies guys! I've used WD Diagnostics, as well as others and the drive passes all!? Even SMART reports all is good. After a fresh format in windows w/the raptor as a slave, I took out the other hdd and disconnected a couple fans, and it finally installed correctly! Now my only complaint is that it's not really that much faster than my WD250KS... hmm... Windows loading time is barely faster, and burst speed is lower than my WD250KS... although average read speed and seek times are both 22mb faster and almost 6ms faster respectively... I'm just not seeing the “speed” as others claim... still debating on whether or not to keep it... What do you guys think?

 

[RichPLS]

now make sure you have all the current SATA controller drivers updated...

 

[Blown_65]

Any help on that... My Gigabyte 965P-S3 has a SATAII/JMicron install that I haven't done yet... but isn't that only for RAID? And... I used the Intel INF auto-installer to install my INF files, because the one's on the CD would crash... BUT it seems to have installed as 965G?? What am I doing wrong? Thanks!!

 

[BustedSony]

Bought a Raptor 74G from Newegg...
Windows could not start because the following file is missing or corrupt:
\Windows\System32\config\system

It did this each time I installed different programs so it wasn't any certain one... plus after the 3rd install of XP some programs wouldn't even load!?
ANYONE?? HELP!! Should I RMA it? Refund? Will a good raptor be worth it? I ran it through a check/scan disk and it came up fine? SMART shows spin up time value as 2300 or something!? HHEEEEEEEELLLPPPPP!!!!

Although you've solved the problem I should mention....
"Windows could not start because the following file is missing or corrupt:
\Windows\System32\config\system"
This message is due to the drive letter changing during install. Maybe what was identified as a C: drive on a previous install is actually the third master in Bios, so when the drive is wiped after loading files into memory but before copying to the drive, the lower-numbered drive in Bios, a data drive, becomes C: and the install or boot drive becomes E: or F:, thus the boot loader can't be found. The trick is to always do a fresh install onto the wanted boot drive with all other drives disconnected.

 

[Blown_65]

That makes perfect sense! I'm such an idiot! That's exactly what must have happend! I'm going to try a fresh install again with the Raptor as the only drive... but first I'll get the ram thing worked out like wusy and woshitudou had mentioned... I'll keep you all posted, I hope this fixes everyting... thanks for all the help guys!!

 

[Blown_65]

Well, I've upped my DDR2 voltages to their "stock" setting @ 2.1V... But it didn't seem to make a difference... I also installed the JMicron/Gigabtye Sata Controller, but no dice their either... The Raptor is fast... just not much faster than my WD250KS... I'm not really dissapointed because it shines at times, but I feel that I was ripped off by WD for their super-light-speed Raptor not being "that" fast... any more suggestions? Thanks again!

 

[BustedSony]

Well, I've upped my DDR2 voltages to their "stock" setting @ 2.1V... But it didn't seem to make a difference... I also installed the JMicron/Gigabtye Sata Controller, but no dice their either... The Raptor is fast... just not much faster than my WD250KS... I'm not really dissapointed because it shines at times, but I feel that I was ripped off by WD for their super-light-speed Raptor not being "that" fast... any more suggestions? Thanks again!

In actual fact the Raptor is NOT that much faster, except for quick access of a lot of small files. You get virtually the same performance quarter-stroking a Seagate 7200.10, 5.3 ms versus 4.5 for the Raptor, and the same throughput. http://blogs.zdnet.com/Ou/?p=322

You can get safe, fast performance, as good as two Raptors in Raid 0, using three, quarter-stroked, Seagates in Raid 5, paying a few dollars more than a single Raptor but getting four times the storage. Buying a Raptor is more about prestige than anything else.

My P4 530 has two Seagate IDE 80 gig drives in Raid 0 on a Sil controller. The C2D E6400 is using one Raptor 74 gig; file access and loading time is very obviously faster on the P4, at one third the cost and with twice the storage.

 

[Blown_65]

Well, I guess since I have it, I might as well use it. It's not bad, I guess my expectations where just really high. Oh well. Really, thanks for your help though!

 

[mkaibear]

>partial stroking

Dear Lord, that article is *rubbish*.

The guy knows *nothing* about what he's talking about.

He's taking the access time as a linear measurement when it's not at all - how much of the time is taken up with accelerating the head to speed, then accelerating it to rest in the right place. Short-stroking a drive might result in a *tiny* performance increase, but it won't "halve or quarter" it like he claims it might.

>RAID 0 vs Raptor
>Perceiving RAID 0 to be faster

And we're back to the old argument. I'm not even going to post the many, many links which say why you are wrong, I'm just going to sigh.

*sigh*

 

[BustedSony]

>partial stroking

Dear Lord, that article is *rubbish*.

The guy knows *nothing* about what he's talking about.

He's taking the access time as a linear measurement when it's not at all - how much of the time is taken up with accelerating the head to speed, then accelerating it to rest in the right place. Short-stroking a drive might result in a *tiny* performance increase, but it won't "halve or quarter" it like he claims it might.

>RAID 0 vs Raptor
>Perceiving RAID 0 to be faster

And we're back to the old argument. I'm not even going to post the many, many links which say why you are wrong, I'm just going to sigh.

*sigh*

Excuse me, but call it what it is, partitioning the drive, and that has always increased performance by using the fastest access part of the platters. There's nothing new in that article, and on my system Sisoft Sandra DOES show almost half the access time a Seagate 7200.10 with one-third partitioned for the boot partition, over using the full drive. The same thing can be determined by looking at benchmarking a drive and seeing the dropoff in transfer speed and increase in access time as the arm moves in.

Of course Raid 0 is faster than a Raptor, you're saying otherwise? But yes in this case it's not access time, which is a bit slower in Raid, but raw transfer speed.

 

[mkaibear]

Yes, you will see an increase in raw sequential transfer rate if you RAID 0 the fastest parts of your drives. You will see a corresponding increase in latency, and in real-world terms you will see no improvement, because the most important thing in real-world terms, unless you're a server, is latency.

Latency trumps raw transfer rate, unless you're doing one of 3 things;

Booting Windows (which *will* be faster on a RAID setup)
Doing HD content creation or *heavy* PS work.
Loading games which *only* require streaming data off the hard drive and don't need the CPU to do any work on it - and there's precious few of these.

If you want the fastest Windows boot times possible, get a RAID 0 setup. To speed up almost all real world tasks, you want a single Raptor.

I am not just blowing this out of my bottom, go and read the anandtech article, or the storagereview article... Or, heck, one of the *many* other articles which support my point of view. I challenge you to find me a single article which supports yours (and by "support", I mean "with a decent range of real-world benchmarks, not synthetic ones")


Just to make things clear - I'm not disputing that you get a *much* faster transfer rate with RAID 0. I'm disputing that this makes the computer itself faster - because it doesn't...

 

[BustedSony]

If you want the fastest Windows boot times possible, get a RAID 0 setup. To speed up almost all real world tasks, you want a single Raptor.

I am not just blowing this out of my bottom, go and read the anandtech article, or the storagereview article... Or, heck, one of the *many* other articles which support my point of view. I challenge you to find me a single article which supports yours (and by "support", I mean "with a decent range of real-world benchmarks, not synthetic ones")


Just to make things clear - I'm not disputing that you get a *much* faster transfer rate with RAID 0. I'm disputing that this makes the computer itself faster - because it doesn't...

I've been doing a lot of reinstalls and upgrades recently to both my Prescott 530 and a C2D E6400 overclocked (easily) to 3.2. gigs. Also I'm running both Vista and XP on the C2D (separate drives/arrays set in Bios hard drive order) and have tried every iteration of Raid 0, Raid 5, and single drives. Partitioned Seagate 7200.10 in Raid 0 is snappier during loading programs and minute to minute use than the single Raptor, or Raided full drives, or a single 7200.10. Unfortunately I couldn't keep a Raid 0 setup on the C2D (I needed space, and there aren't enough serial slots or space for multiple drives in the P180) but on then other hand the Prescott has two Seagate IDE 80 gig drives in Raid 0, unpartitioned, and I can't tell you how fast and responsive the system is. It's ideal for the audio work and DVD authoring jobs it was built for.

The Prescott, for going from one thing to the next, is faster than the C2D, because of the different drive setups.

Naturally the C2D floors the Prescott for rendering speed. I can say now the C2D, an E6400 at 3.2 gig, is OVER twice the speed in DivX and HC encoding than the P530 overclocked to 3.6 gHz, and a lot lot cooler.

One doesn't need synthetic benchmarks or questionable articles when one just KNOWS what works! For running Oblivion loading areas works best with a Single Raptor, just beating out the Raid 0 7200.10. Just. This is probably where raw access speed matters most. For everything else the Raid 0 setup has less waits for action (opening Explorer, starting a program) than the Raptor.

I know the virtues of the Raptor, but they have been exaggerated by those looking to justify the higher price.

 

[mkaibear]

>one doesn't need synthetic benchmarks or questionable articles when one just KNOWS what works!

Hah.

So, basically, you're saying "perception is better than science".

I can show you benchmarks that scientifically, and repeatably, show that a RAID implementation is *not* as fast in general desktop use than a single Raptor.

I can show you benchmarks that scientifically and repeatably, show that a RAID implementation with anything other than a hardware RAID card is *not* as fast in general desktop use than a single drive.

You can not show me any benchmarks, scientific or otherwise, to show me anything other than the fact that a RAID setup has a higher STR than a non-RAID setup.

You show me proof, I'll believe you. In the meantime, I guess you'll just have to stick with the "I feel this to be the case, therefore it is the case" attitude you have at the moment.

 

[BustedSony]

>one doesn't need synthetic benchmarks or questionable articles when one just KNOWS what works!

Hah.

So, basically, you're saying "perception is better than science".

You can not show me any benchmarks, scientific or otherwise, to show me anything other than the fact that a RAID setup has a higher STR than a non-RAID setup.

You show me proof, I'll believe you. In the meantime, I guess you'll just have to stick with the "I feel this to be the case, therefore it is the case" attitude you have at the moment.

I'm not going to get into a pointless argument. The OP said he was disappointed that using a Raptor did not seem to give as big a boost over a regular drive as he expected. I agreed with him because that was my own experience also, and suggested a link he should visit. You come along and say Benchmarks show such and such. I'm not talking about Benchmarks, I'm talking about day to day use. That's all. Likewise the speed up in encoding speed of the C2D is not a "benchmark" but actual use.

And yes I know about Office productivity Benchmarks, please give me a link to one that compares a short-stroked 7200.10 Raid 0 versus a single Raptor. Please. Since that is the scenario I'm using.

 

[mkaibear]

I can show you many which prove that a non short-stroked RAID 0 can't compete with a Raptor in day to day usage.

The only place I've ever seen anyone talk about such an odd idea as short-stroking is that link you posted - and it looks like a load of rubbish to me, the guy doesn't know what he's talking about.

Please can you provide me some benchmarks (you know, the "scientific" kind, not just "I think it's better, therefore it's better") showing that a short-stroked RAID 0 will perform better than a Raptor, in *anything* other than Sequential Transfer Rate?

 

[BustedSony]

I can show you many which prove that a non short-stroked RAID 0 can't compete with a Raptor in day to day usage.

The only place I've ever seen anyone talk about such an odd idea as short-stroking is that link you posted - and it looks like a load of rubbish to me, the guy doesn't know what he's talking about.

Please can you provide me some benchmarks (you know, the "scientific" kind, not just "I think it's better, therefore it's better") showing that a short-stroked RAID 0 will perform better than a Raptor, in *anything* other than Sequential Transfer Rate?

I notice that you're riding the same horse in this thread, http://forumz.tomshardware.com/hardware/RAID-improves-writting-speed-reading-speed-ftopict205641.html Forget what users experience day to day, it's BENCHMARKS that count .. I hadn't read that thread before yet it's interesting that I've been saying the same thing here, that boot loading times and general data transfer is quicker with Raid 0, but game loading times are no different. So I'm not alone thinking this, you are in what you think.

http://www.overclockers.com/articles1063/index02.asp

Quote "As to how Raid-0 stacks up against a single Raptor, the only difference is in the benchmark scores - other then that, in real world use there is NO REAL IMPROVEMENT in load up times.
Conrad aka Vio1 in OC Forum "
-thus suggesting that Benchmarks don't tell the full story, and that they would favour my viewpoint



As for short-stroking/partitioning the boot drive, this is a no-brainer. You are confining the files to a small part of the disc, therefore access will be quicker because the head doesn't have to travel all over the disc, AND since all desktop drives have a falling curve on transfer speed you will be using the fastest part of the disc. Of COURSE short-stroking is quicker! And I wish there WERE a benchmark somewhere.


Now that my two systems are working and in use I'm sure as Hell not going to tear them apart and start a Benchmark series. However one thing that DID happen is with the Windows Vista experience index. Full single 7200.10 was 5.6, short-stroking was 5.7, Raid 0 Short resulted in 5.8, and yes the Raptor 74 gave 5.9. So Vista likes the Raptor.