//flex table opened by JP

Click to See Complete Forum and Search --> : IBM Reveals Tag 'N Seek Technology


Bsamuels
10-02-2002, 02:31 PM
hopefully it wont burn up if u let it run for 20 mins :rolleyes:

Aarmenaa
10-03-2002, 08:15 PM
How nice of them to put tags on everything, so you can identify your dead files afterwards...kinda like the tags they put on the dead after a disaster...hmmm. :r

-Aarmenaa

corrupted
10-03-2002, 08:51 PM
it sounds like they're merging SCSI and IDE. we needed a new format anyway.

Tuclis
10-04-2002, 09:55 AM
I thought IBM got out of the HD business... or was it just the manufacturing of the HDs?

corrupted
10-05-2002, 08:35 PM
there was talk, but then they realised where all their money was really going. let's just say that employees had a REALLY good deal till they found out.

now they manufacture drives, but on a much smaller scale. just enough for themselves, and the spares get sent out to warehouses so that we mortals can buy them.

bottom4s
10-07-2002, 09:44 AM
What? Are they talking about the IMB DEATHSTAR? Don't waste your money!!!

dosmastr
10-07-2002, 03:48 PM
I know the sting of dead drives, i have 2 IBM 75GXP's and i've RMAed them both 2X, one is about to go in for time #3. and while the tag the dead comment is a pretty good one. i have to let you all know, that only the 75GXP and some early 60GXP's had the problems that ibm is being sued for, the 120GXP was a totally solid drive, nothing against it on the reliability front.

how many commands long is the queue? a full 256 like SCSI or (lol) just 2 or 4?

how can they benchmark a drive like that without using a SCSI benchmark? wouldn't an IDE benchmark simply give it only 1 command at a time forcing it to function as a normal drive?
what am i missing here?

corrupted
10-07-2002, 07:09 PM
i think you're right dosmaster, only a SCSI benchmark will pick up the IDE hardware and refuse to operate. tricky huh?

thronka
10-07-2002, 10:50 PM
The only operating system that supports the Tag'n'Seek technology is BeOS 5. Windows XP doesn't support Tag'n'Seek.

corrupted
10-07-2002, 11:08 PM
i'm sure they'd have drivers available, though it shouldn't make any difference to the performance, should it?

thronka
10-07-2002, 11:14 PM
Looks like with the ATA limitations that I would say no. Remember, the Chipset really determines speed.

corrupted
10-07-2002, 11:24 PM
good point. we'll need to wait for the mobo chipset makers to incorporate support before we can fully use the advantages this provides.

thronka
10-07-2002, 11:55 PM
I would like to see, more than anything, is better access times and lower heat. That, more than anything, would increase speed up times no metter what chipset.

corrupted
10-08-2002, 12:59 AM
i've been thinking about racks for mine because of that. pity the laptop can't have one fitted. :(

bottom4s
10-08-2002, 12:28 PM
Sorry Dosmaster, but mine is a 60gig. I purchased it less than 9months ago. It was manufactured in Oct. 2001. Now, that's what you call earlier. Then mine is one of the earlier ones. I just replaced it with another brand that hasn't exhibited these failures. By the way, I also sent my drive back for RMA in the replacement brand's box:) Let's make a statement.

edwardng
10-08-2002, 01:35 PM
Which model series was your 60GB Deskstar? 75GXP, 60GXP or 120GXP? One of my 15GB 75GXPs died recently in my RAID array; since I had a 400BB supplemental drive, I just bought another 400BB and set up a new array; the remaining good 75GXP is my supplemental drive and the toasty 75GXP sits on my desk. Nice heavy paperweight.

jad1097
10-08-2002, 02:21 PM
Originally posted by dosmastr
I know the sting of dead drives, i have 2 IBM 75GXP's and i've RMAed them both 2X, one is about to go in for time #3. and while the tag the dead comment is a pretty good one. i have to let you all know, that only the 75GXP and some early 60GXP's had the problems that ibm is being sued for, the 120GXP was a totally solid drive, nothing against it on the reliability front.

how many commands long is the queue? a full 256 like SCSI or (lol) just 2 or 4?

how can they benchmark a drive like that without using a SCSI benchmark? wouldn't an IDE benchmark simply give it only 1 command at a time forcing it to function as a normal drive?
what am i missing here?

You are incorrect.

http://www.tech-report.com/news_reply.x/2799/

http://www.sysopt.com/forum/showthread.php?s=&threadid=109276