heated snail

Member
  • Content Count

    181
  • Joined

  • Last visited

Community Reputation

0 Neutral

About heated snail

  • Rank
    Member
  1. I was going to bring up the massive speed increase available with Redundant Arrays of Inexpensive Cardstock, but I'm sure it's been discussed here at length already.
  2. Think of all the jobs created for the army of data librarians lugging your terabyte of punchcards around campus on wheeled carts. And hey, the librarians might even be cute.
  3. heated snail

    IBM 60GXP/75GXP warranty replacemenet drives

    My latest 75GXP that died (a month ago) was replaced with a refurbished 75GXP that failed DFT upon arrival. I was able to get them to pay for the return shipping but they would not (on the phone) guarantee a 120GXP in exchange -- mainly, they said, because the telephone-tech support/RMA department are not in contact with the people who actually receive/test/replace the drives, who may or may not have the requested replacement drive on hand at the time. Meanwhile a 60GXP died on me so in effect I'm asking for 120GXP (minimum) replacements for both. Which I will probably still sell on arrival, assuming they pass DFT. Now I'm crossing my fingers hoping that none of my WD800JBs develop bearing noise.
  4. heated snail

    W2K + SCSI = Performance problems for me

    Avenger, if you only have one drive in your system that's the #1 problem for Photoshop performance. One of the most effective ways to achieve faster results in Photoshop is to have multiple drives that you can assign as Photoshop "scratch disks". Having the image you're editing sourced and saved on a drive other than your system drive is a good idea too, because then image loading and saving won't have to be affected by virtual memory paging to disk (assuming that's happening on your system drive). If you have enough drives you can split your pagefile across several drives and split your Photoshop temporary files across several OTHER disks and thereby have the fewest possible interferences between different operations.
  5. archonix, you'd see better bootup and program-startup times from the 120GXP array due to IBM's faster seek times. Meanwhile the WD drives are faster on moving large amounts of data. If you insist on booting from a RAID array then the 120GXPs would be the better choice, while the WDs make a better array for storing large data files, like audio/visual stuff. Having said that, if I had to pick ONE of the above drives to boot off of, not an array, I'd still choose the WD and sell the IBMs as soon as possible. 120GXPs are more reliable than 60GXPs and far more reliable than 75GXPs, but they're not beyond suspicion. Mine have made unpleasant noises on occasion. Meanwhile WDs seem to either have problems right after you buy them, or not at all -- though the relevant behavior of the new JB series remains to be seen over time.
  6. Yep, on second thought "Free to try" really doesn't fit the Free criterion. But I thought you might be curious to see it.
  7. Vopt... It's free for a month, so I don't know if it fits your free criteria. They have versions for DOS 3.1, NT4, 9x, 2K, XP. It's mighty fast but I have no idea how its safety/integrity compares to others, nor am I aware of whether it uses Windows' preferred file ordering when optimizing. It includes some extra tools besides the defragger, some of which are only interfaces to Windows capabilities, but some are useful additions like a pagefile mover/defragger. www.vopt.com
  8. ER, yes, there's a tool for just that purpose. You can boot off the Windows XP CD into an environment called 'Recovery Console' which is a DOS-like environment that can access NTFS partitions. I'm not experienced enough to say more about it in general though I have used it a few times to repair and replace corrupted registry hives and boot loaders. Look up Recovery Console via Microsoft Help and you'll find a wealth of information about it.
  9. sechs, thanks for the correction. Instead of "a new drive should have no remapped sectors" I should have said it shouldn't have very many. Jase, as you can see Winbench seems to get around system interruptions (plus you said you already paid some attention to cleaning up Windows processes) but there are still those couple of obvious spikes. Those could certainly be remapped sectors but it's hardly anything to worry about.
  10. heated snail

    Terrible SCSI performance in Windows XP

    fa147, I installed Adaptec's SP4 U160 drivers when they first went "final" around November 16th, and haven't noticed anything better or worse in my system performance. However, I currently have only one hard drive attached to my 29160N, plus a CD burner.
  11. heated snail

    W2K + SCSI = Performance problems for me

    Avenger, according to the SCSI Performance in XP thread, Win2K SCSI performance is just as poor if you install SP3. Are you on SP3, or an earlier version of 2000?
  12. Jase, if there were just a few of those spikes it could mean remapped sectors... hard drives are provided with a batch of "extra" sectors that can be used if necessary... when the drive notices it has a sector going bad it can replace that sector with one of those "extra" sectors. However, after this is done, there's now a sector that's in line with the rest logically/virtually, but way off somewhere else on the drive physically. So if a file that is defragmented/contiguous happens to use one of the remapped sectors, when that file is read the drive will have to make a big seek in the middle of the file to the remapped area and back again. The same effect can cause a big dip in a STR graph like that. But a new drive should have no remapped sectors for a long time. And when the graph shows a great many dips like yours does, it's usually just that your processor and IDE bus are getting interrupted briefly in the middle of the test. Probably by a combination of various Windows services and any other background processes you may have left running. For example if you're on Windows 2K or XP and the "indexing service" is left "on" for any drive in your system, it can easily interrupt and prevent a smooth transfer-rate graph. So can XP's System Restore function. There are a large number of other potential interruptions and generally you aren't free of them all until you spend some time carefully turning off all Windows services that you don't really need. See www.blackviper.com for an example of one man's very useful guide to what services can be turned off and how safely. Beyond that I can't give more suggestions since you didn't mention what operating system you're running.
  13. heated snail

    Is this a good score for a raid 0

    Yeah, that's a great score, for 2 IDE drives in RAID 0, assuming you benched the entire drive and it was empty. Note, however, that HDTach doesn't do the most reliable job of testing RAID arrays. You can get more feedback if you post a Winbench 99 "drive inspection test" transfer rate graph for the entire array. In the future it's a good idea to mention explicitly what drives you benched with, rather than leaving us to assume that the 2 listed below in your rig are the ones you used with the 404.
  14. Far out. I sense short term history in the making.
  15. I just replaced my failing 60GXP with a WD800JB. 3 year warranty. WD800JB at Fry's Electronics (also Outpost.com): $140, with a $40 mail-in rebate. WD600JB at Fry's: $100. Might as well buy a 80 for the price of 60. DM+9 80Gb at Fry's: $140 but no rebate.