🚀 go-pugleaf

In article <46hcd8$t18@elaine16.Stanford.EDU>, kla@leland.Stanford.EDU (Earl Mitchell) writes:
>
>I'm looking for good reliable 4GB SCSI drive for my PC.
>What should I go for the Seagate Barracuda or Quantum Grand PRix?
>The Quantums appear significantly cheaper but are they better
>or as good as the Seagates? Better meaning more reliable.

I haven't heard much good about the Barracudas.
They tend to die unexpectedly and often, from what I hear.
Just my opinion.

On 24 Oct 1995 02:29:45 GMT, vp24njcb@ubvms.cc.buffalo.edu <vp24njcb@ubvms.cc.buffalo.edu>
wrote:
 : >I'm looking for good reliable 4GB SCSI drive for my PC.
 : >What should I go for the Seagate Barracuda or Quantum Grand PRix?
 : >The Quantums appear significantly cheaper but are they better
 : >or as good as the Seagates? Better meaning more reliable.
 : 
 : I haven't heard much good about the Barracudas.
 : They tend to die unexpectedly and often, from what I hear.
 : Just my opinion.

Micron systems come with 4 Gig Connor drives.  Are they any good?  I changed
the subject line to include the Connor drives.

Thanks,
--John

vp24njcb@ubvms.cc.buffalo.edu wrote:
>In article <46hcd8$t18@elaine16.Stanford.EDU>, kla@leland.Stanford.EDU (Earl Mitchell) writes:
>>
>>What should I go for the Seagate Barracuda or Quantum Grand PRix?
>>The Quantums appear significantly cheaper but are they better
>>or as good as the Seagates? Better meaning more reliable.
>
>I haven't heard much good about the Barracudas.
>They tend to die unexpectedly and often, from what I hear.

If you don't cool them well either will fail quickly.
If you cool them properly, either will perform reliably.

It's really that simple. An ordinary desktop PC does not usually
have enough airflow to cool high performance drives sufficiently.

The Barracuda will have a slightly higher performance than the Grand Prix.
Price/performance depends on your source, of course.

Hint: If you get a noisy Grand Prix just return it. We fixed the noise
issue some time ago but there are probably still a lot of older drives
around that just have to wait until they end up in noise insensitive
environments.

Ralf-Peter
========================================================================
Ralf-Peter Rohbeck                                     rrohbeck@qntm.com
Quantum GmbH-Application Engineering-Central Europe   (+49) 69-950767-18
Berner Str. 28, 60437 Frankfurt, Germany          fax (+49) 69-950767-91
#include "disclaimer.h"          tech support hotline (+49) 69-950767-26

In article <46hj2p$r6t@azure.acsu.buffalo.edu>, vp24njcb@ubvms.cc.buffalo.edu says...
>
>In article <46hcd8$t18@elaine16.Stanford.EDU>, kla@leland.Stanford.EDU (Earl Mitchell) writes:
>>
>>I'm looking for good reliable 4GB SCSI drive for my PC.
>>What should I go for the Seagate Barracuda or Quantum Grand PRix?
>>The Quantums appear significantly cheaper but are they better
>>or as good as the Seagates? Better meaning more reliable.


lately i've been reccomending the Fujitsu or the Conner
the Fujitsu has the new MR heads & is a little cheaper and
the Conner is a little faster
both are great AV drives


SCSI peripherals  /  WHOLESALE    grasser@srmc.com
                                  http://sweb.srmc.com/andys/index.html
                                  Voice 310-546-3033 Fax 310-546-2013
               ___                 
 __.--/)  .-~~   ~~>>>>>>>>   .-. Disk Drives 1-9gb / RAID 1gb-100gb
(._\~  \ (        ~~>>>>>>>>.~.-' Seagate/Fujitsu/Hitachi/Conner/Quantum
  -~}   \_~-,    )~~>>>>>>>' /    Tape Drives 4mm,8mm,1/4",1/2",DLT
    {     ~/    /~~~~~~. _.-~     Exabyte/Sony/Conner/Tandbrg/OD/Quantum
     ~.(   `--~~/      /~ ~.      CD-ROM/CD Recorders/M/O Optical/Robots
.--~~~~_\  \--~(   -.-~~-.  \     ONLY new w/ full mfg's warranty
```-'~~ /  /    ~-.  \ .--~ /
     (((_.'    (((__.' ```-'      * chocolate isN't good for dogs 
 

fishman@panix.com (Harvey Fishman) wrote:
>In <46hj2p$r6t@azure.acsu.buffalo.edu> vp24njcb@ubvms.cc.buffalo.edu says

>> I haven't heard much good about the Barracudas.
>> They tend to die unexpectedly and often, from what I hear.
>> Just my opinion.

>I have had an ST-15150N in constant operation since April with no
>problems.  I have a friend with very similar experiences.  My ISP has a
>whole passel of them in constant operation for a much longer period of
>time without inordinate failure rates.  The important thing here is to be
>meticulous about the cooling.  The major problem is that the people who
>are too lazy to read and follow the manufacturer's instructions are
>usually the ones with the biggest and baddest mouths.  But I guess that
>that is a consistant pattern of arrogance.


Harvey Fishman is the poster boy for arrogance.

Followers of these groups have read many reports of the Barracuda
problems -- perhaps the drives are no longer so unreliable, I bet the
new ones are better.  We all know cooling these hot 7200 drives is of
course mandatory.  

But only the poster boy of arrogance would presume to call the lot of
them "lazy" and suggest they all didn't follow proper cooling
guidelines.    Many of people complaining were from large operations
with many, many drives -- with extensive knowledge and experience,

just another view...

In <46hj2p$r6t@azure.acsu.buffalo.edu> vp24njcb@ubvms.cc.buffalo.edu says

> In article <46hcd8$t18@elaine16.Stanford.EDU>, kla@leland.Stanford.EDU (Earl Mitchell) writes:
> >
> >I'm looking for good reliable 4GB SCSI drive for my PC.
> >What should I go for the Seagate Barracuda or Quantum Grand PRix?
> >The Quantums appear significantly cheaper but are they better
> >or as good as the Seagates? Better meaning more reliable.
> 
> I haven't heard much good about the Barracudas.
> They tend to die unexpectedly and often, from what I hear.
> Just my opinion.

I have had an ST-15150N in constant operation since April with no
problems.  I have a friend with very similar experiences.  My ISP has a
whole passel of them in constant operation for a much longer period of
time without inordinate failure rates.  The important thing here is to be
meticulous about the cooling.  The major problem is that the people who
are too lazy to read and follow the manufacturer's instructions are
usually the ones with the biggest and baddest mouths.  But I guess that
that is a consistant pattern of arrogance.

But seriously, if you are unwilling to devote the time and $$$ to do the
proper cooling then definitely avoid Barracudas.  But if you give them the
proper care, feeding, and love they will serve you long, well, and fast.

Harvey

--
----------------------------------------------------------------------------
 Harvey Fishman   |   It is easy to live by the laws that you agree with.
fishman@panix.com |          But being law-abiding means obeying
  718-258-7276    |        even the ones that you think are wrong.


                                                            

In article <46hj2p$r6t@azure.acsu.buffalo.edu> vp24njcb@ubvms.cc.buffalo.edu writes:
>In article <46hcd8$t18@elaine16.Stanford.EDU>, kla@leland.Stanford.EDU (Earl Mitchell) writes:
>>
>>I'm looking for good reliable 4GB SCSI drive for my PC.
>>What should I go for the Seagate Barracuda or Quantum Grand PRix?
>>The Quantums appear significantly cheaper but are they better
>>or as good as the Seagates? Better meaning more reliable.
>
>I haven't heard much good about the Barracudas.
>They tend to die unexpectedly and often, from what I hear.
>Just my opinion.

   NewMedia magazine did a test of 14 High Perf. SCSI drives in
June '95.  They rated the Barracudas higher in performance.  I'm
not sure about dying "unexpectedly and often".  I rarely see a
problem that is actually the drive's fault.  I would suggest
also looking into the Connor Filepro 4207. Connor has a very
good reputation.  I think that drive is in the same price range.

		Mark.

In article <46j2r1$bh4@ixnews4.ix.netcom.com>,
grasser <grasser@srmc.com> wrote:
>In article <46hj2p$r6t@azure.acsu.buffalo.edu>, vp24njcb@ubvms.cc.buffalo.edu says...
>>
>>In article <46hcd8$t18@elaine16.Stanford.EDU>, kla@leland.Stanford.EDU (Earl Mitchell) writes:
>>>
>>>I'm looking for good reliable 4GB SCSI drive for my PC.
>>>What should I go for the Seagate Barracuda or Quantum Grand PRix?
>>>The Quantums appear significantly cheaper but are they better
>>>or as good as the Seagates? Better meaning more reliable.
>
>
>lately i've been reccomending the Fujitsu or the Conner
>the Fujitsu has the new MR heads & is a little cheaper and
>the Conner is a little faster

Yeah, on reads, I'd believe that.  The Conner can read at nearly 7MB/sec 
off the _slow_ edge of the disk.  But its write performance is abysmal, 
and units are still working their way through distribution which have 
tremendous stiction problems and often need to whacked really hard 
before they'll spin up.

I would not recommend the Conner.  In the same price range, I'd 
recommend the Quantum Grand Prix.  I am stuck with 50 of the Conners in a 
critical application, and believe you me, I know a lot more about them 
than I'd like to.

Early Fujitsu MR drives had a lot of problems.  They probably do have 
them licked by now.  But if you don't mind that it's loud as the Last 
Trump, the Quantum can't be beat.  And its Atlas high-performance brother 
is simply the very best.
-- 
Thor Lancelot Simon                                               tls@cloud9.net
 
Don't let your mouth write no check that your tail can't cash.      --Bo Diddley

davis@space.mit.edu (John E. Davis) writes:

>On 24 Oct 1995 02:29:45 GMT, vp24njcb@ubvms.cc.buffalo.edu <vp24njcb@ubvms.cc.buffalo.edu>
>wrote:
> : >I'm looking for good reliable 4GB SCSI drive for my PC.
> : >What should I go for the Seagate Barracuda or Quantum Grand PRix?
> : >The Quantums appear significantly cheaper but are they better
> : >or as good as the Seagates? Better meaning more reliable.
> : 
> : I haven't heard much good about the Barracudas.
> : They tend to die unexpectedly and often, from what I hear.
> : Just my opinion.

Barracudas tend to run HOT!  Give them lots of airflow, and they behave 
nicely.


Gerald

kaye@aloha.com wrote:

> Harvey Fishman is the poster boy for arrogance.

     I've been following this group (comp.periphs.scsi) for quite some
time, and I would say that the "poster boy" label would fit you better
than Harvey.  I have not seen him making any personal attacks.  While I
don't necessarily agree with Harvey, I also don't "name call" when I
disagree.  And, if I really "disagree", I just enter them into my kill
file, which currently has 90+ entries ....

> Followers of these groups have read many reports of the Barracuda
> problems -- perhaps the drives are no longer so unreliable, I bet the
> new ones are better.  We all know cooling these hot 7200 drives is of
> course mandatory.  

     Yes, and, of the postings that I recall, many of them mentioned
that the dead/dying Barracuda was "too hot to touch".

     I've had a 12550N Barracuda since June/July of last year, and it's
been running fine.  However, I've been very careful to keep it *cool*.
I would be willing to admit that there *might* be a problem with the
32550 Barracuda, though.

     As a side note, I'm currently toying with getting a 4GB drive, and
the 15550 Barracuda is one of the drives that I'm considering ....

     -- Darryl Okahata
	Internet: darrylo@sr.hp.com

DISCLAIMER: this message is the author's personal opinion and does not
constitute the support, opinion or policy of Hewlett-Packard or of the
little green men that have been following him all day.

In article <46m88h$53@news.aloha.com>, kaye@aloha.com writes:
> Followers of these groups have read many reports of the Barracuda
> problems -- perhaps the drives are no longer so unreliable, I bet the
> new ones are better.  We all know cooling these hot 7200 drives is of
> course mandatory.  

  I've had ST12550N's since revision 10, and ST32550N's since revision 9
(pre-production) and while I've had a few drive failures, overall I've
been very happy with the product. I've had excellent experiences with
Seagate - both at the local level and at the corporate level (my local
rep involved the VP for Quality Assurance when I raised an issue). Many
of my offhand suggestions have actually been implemented and are now in
the field. And I'm just an end user, buying maybe a few hundred drives a
year, all from various distributors.

> But only the poster boy of arrogance would presume to call the lot of
> them "lazy" and suggest they all didn't follow proper cooling
> guidelines.    Many of people complaining were from large operations
> with many, many drives -- with extensive knowledge and experience,

  I've seen a lot of baked drives, not just Seagate ones. Looking at some
magazine ads, I see a lot of "big players" doing the "slap it in a case
and ship it". The detailed Seagate product manuals are free for the asking -
either do your own research or find somebody dependable to do it for you.

  Size is no guarantee that things are done right.

	Terry Kennedy		  Operations Manager, Academic Computing
	terry@spcvxa.spc.edu	  St. Peter's College, Jersey City, NJ USA
        +1 201 915 9381 (voice)   +1 201 435-3662 (FAX)

In article <46o8po$f9t@ixnews7.ix.netcom.com>,
grasser <grasser@srmc.com> wrote:
>In article <46kldv$4ma@news.cloud9.net>, tls@cloud9.net says...
>>
>>In article <46j2r1$bh4@ixnews4.ix.netcom.com>,
>>grasser <grasser@srmc.com> wrote:
>>>In article <46hj2p$r6t@azure.acsu.buffalo.edu>, vp24njcb@ubvms.cc.buffalo.edu says...
>>>>
>>>>In article <46hcd8$t18@elaine16.Stanford.EDU>, kla@leland.Stanford.EDU (Earl Mitchell) writes:
>>>>>
>>>>>I'm looking for good reliable 4GB SCSI drive for my PC.
>>>>>What should I go for the Seagate Barracuda or Quantum Grand PRix?
>>>>>The Quantums appear significantly cheaper but are they better
>>>>>or as good as the Seagates? Better meaning more reliable.
>>>
>>>
>>>lately i've been reccomending the Fujitsu or the Conner
>>>the Fujitsu has the new MR heads & is a little cheaper and
>>>the Conner is a little faster
>>
>>Yeah, on reads, I'd believe that.  The Conner can read at nearly 7MB/sec 
>>off the _slow_ edge of the disk.  But its write performance is abysmal, 
>>
>>I would not recommend the Conner.  In the same price range, I'd 
>>recommend the Quantum Grand Prix.  I am stuck with 50 of the Conners in a 
>>critical application, and believe you me, I know a lot more about them 
>>than I'd like to.
>
>i don't think so
>1. more people have complained about the Grand Prix's loudness than 
>   any drive i've ever sold
>2. i had more DOA (includes within the 1st month) Grand Prixs than 
>   any drive i've ever sold
>3. if i'm not mistaken, the Grand Pric has been discontinued 
>   in favor of the Atlas (which isn't as fast as the Conner 4207
>   even though the specs are better)

Have you actually benchmarked this extensively, or are you just saying 
that they're faster because you think so?

I don't believe the Grand Prix has been discontinued.

I have 50 Conner 4207s here, all in the same room, all essentially brand 
new, all with the _exact same_ stiction problem, which is infuriating.

They do perform quite nicely on reads, around 6.8MB/sec.  If all you ever 
use your disk for is reads, I suppose that's great.  But they only write 
at about 3.6MB/sec (all on long sustained writes), so they essentially lose,
in most real-world applications.

I have four units which have already exhibited more than five new bad blocks.

They are _not_ from the same production run.

As for the Grand Prix drives, they don't read quite as fast as the 4207 
(6.3-6.5MB/sec) but they write essentially as fast as they read, which 
means that they have exceptional write performance.  I dont' have as many 
as I do of the Conners, but I haven't seen one fail yet.  They really are 
extraordinarily loud, but I actually intend to _use_ them, not _listen_ 
to them, so I could care less.

I don't believe the Grand Prix has been discontinued, but I am interested 
by your assertion that the 4207 is faster than the Atlas; it certainly 
isn't on writes, and while I haven't extensively tested the Atlas, 
preliminary tests with the 4GB model show over 9MB/sec on many long reads,
a full 2MB/sec faster than the 4207 under the same conditions.

-- 
Thor Lancelot Simon                                               tls@cloud9.net
 
Don't let your mouth write no check that your tail can't cash.      --Bo Diddley

In article <46oj7p$bs1@news.cloud9.net> tls@cloud9.net (Thor Lancelot Simon) writes:

  Have you actually benchmarked this extensively, or are you just saying 
  that they're faster because you think so?

  I don't believe the Grand Prix has been discontinued.

  I have 50 Conner 4207s here, all in the same room, all essentially brand 
  new, all with the _exact same_ stiction problem, which is infuriating.

  They do perform quite nicely on reads, around 6.8MB/sec.  If all you ever 
  use your disk for is reads, I suppose that's great.  But they only write 
  at about 3.6MB/sec (all on long sustained writes), so they essentially lose,
  in most real-world applications.

Well, I can't speak to stiction problems, but in my testing of sustained
writes to a Conner 4207 I achieved a sustained 6384024 bytes/second to the
/dev/sdb device under Linux.  What was your test environment?

		Leonard

tls@cloud9.net (Thor Lancelot Simon) writes:

>                          [Quantum Grand Prix]  They really are 
>extraordinarily loud, but I actually intend to _use_ them, not _listen_ 
>to them, so I could care less.

There is specification deficiency in the earlier model Grand Prix drives 
that has been corrected. If the noise level is bothersome, direct 
inquiries to Quantum or the Quantum authorised dealer concerning replacement.


>I don't believe the Grand Prix has been discontinued, but I am interested 
>by your assertion that the 4207 is faster than the Atlas; it certainly 
>isn't on writes, and while I haven't extensively tested the Atlas, 
>preliminary tests with the 4GB model show over 9MB/sec on many long reads,
>a full 2MB/sec faster than the 4207 under the same conditions.

I've done extensive testing of Atlas drives and found them to be quite 
fast, I'm looking forward to benchmarking the 4207 series and will add 
the results to what I've made available via FTP at ftp.rahul.net in 
/pub/lps/hard-disk/performance/

The Hewlett Packard SureStore series also look competitive, I've just 
purchased several and I haven't seen any other drives that come with as 
much documentation or accessories for installation.
-- 
------------------------------------------------------------------------
 Kevin Martinez                 |           Fry's Electronics: Where
 lps@rahul.net                  |         Incompetence is the Standard!
------------------------------------------------------------------------

tls@cloud9.net (Thor Lancelot Simon) wrote:

>They really are 
>extraordinarily loud, but I actually intend to _use_ them, not _listen_ 
>to them, so I could care less.

I'm looking for a quiet AV SCSI drive because I am going to use it in
a Digital Audio Workstation (DAW) in a studio. Noise is very
important. I have already ruled out the Micropolis2 & 4 GB units for
this reason. Can you guys give me some info on the Fujitsu, Barracuda
and Atlas relative to noise and heat? The heat is an issue if I need
to add additional fans. Thanks



______
Keith Hollister
keithh@gate.net

Darryl Okahata writes:

 >      I've had a 12550N Barracuda since June/July of last year, and it's
 > been running fine.  However, I've been very careful to keep it *cool*.
 > I would be willing to admit that there *might* be a problem with the
 > 32550 Barracuda, though.

   I've had some experience with a few Seagate drives recently.

   I've had 120 ST32550W (2.1 GB) Barracudas running almost
continuously since July of this year, with no failures.  They are well
cooled, though.

   I've also had 72 ST410800W (9 GB) Elites running for a few months,
with 2 of them giving out early in life.

   I think I'm starting to have problems with a ST31230N Hawk in a
poorly cooled bay in my Linux machine (and I believe I lost a ST31200
in the same location due to heat).

-- 
David Myers                                   dem@skyline.rtp.nc.us
Cary, N.C.

In article <46otjg$1s9k@news.gate.net>,
Keith Hollister <keithh@gate.net> wrote:
>tls@cloud9.net (Thor Lancelot Simon) wrote:
>
>>They really are 
>>extraordinarily loud, but I actually intend to _use_ them, not _listen_ 
>>to them, so I could care less.
>
>I'm looking for a quiet AV SCSI drive because I am going to use it in
>a Digital Audio Workstation (DAW) in a studio. Noise is very
>important. I have already ruled out the Micropolis2 & 4 GB units for
>this reason. Can you guys give me some info on the Fujitsu, Barracuda
>and Atlas relative to noise and heat? The heat is an issue if I need
>to add additional fans. Thanks

The Atlases I have seem pretty quiet, and reasonably cool.  The Ultrastar 
DFHSes I have are quiet, but run rather hot.  I don't know about Fujitsu 
drives in that regard; I have a few 32550N 2GB Barracudas which run 
(modulo being in rackmount cases with huge huge fans on both sides) very 
cool, and are quiet enough that at least over the fan noise (not really 
so bad, with the case in question) I can't hear them seek.  For some 
reason, though, I always get lousy write performance with the Barracudas; 
I've noticed this several times.  Does anyone know if there's a mode page 
bit that might be mis-set for many applications by default?
-- 
Thor Lancelot Simon                                               tls@cloud9.net
 
Don't let your mouth write no check that your tail can't cash.      --Bo Diddley

vp24njcb@ubvms.cc.buffalo.edu wrote:

>In article <46hcd8$t18@elaine16.Stanford.EDU>, kla@leland.Stanford.EDU (Earl Mitchell) writes:
>>
>>I'm looking for good reliable 4GB SCSI drive for my PC.
>>What should I go for the Seagate Barracuda or Quantum Grand PRix?
>>The Quantums appear significantly cheaper but are they better
>>or as good as the Seagates? Better meaning more reliable.

>I haven't heard much good about the Barracudas.
>They tend to die unexpectedly and often, from what I hear.
>Just my opinion.

We have 3 Barracudas in our Novell server and had to install another
fan to blow directly on them to cool them down otherwise they tended
to die when put under a heavy load (nightly backups).   No problems
since I put the fan on them though.

Hello, ...
Thor Lancelot Simon (tls@cloud9.net) wrote:
: The Ultrastar 
: DFHSes I have are quiet, but run rather hot.  

my 4.3 GB DFHS is passable quiet; I don't hear it until it comes to
excessive filesystem updates. The drive steps really fast but not very quiet.

Concerning temperature I have to say, that I don't it runs very hot, but
rather some kind of 'warm'. My Fujitsu 2694 (1Gig, 5400RPM) runs remarkable
hotter.
(ahc0:1:0): "IBM OEM DFHSS4F 4040" type 0 fixed SCSI 2
sd1(ahc0:1:0): Direct-Access 4303MB (8813870 512 byte sectors)
sd1(ahc0:1:0): with 4392 cyls, 16 heads, and an average 125 sectors/track

Greetings,
   Andreas

--
       +------------------------------------------------------+
       | Andreas Braukmann             University of Dortmund |
       +------------------------------------------------------+
       |       braukman@ls12.informatik.uni-dortmund.de       |
       |       ab@ltd2.maschinenbau.uni-dortmund.de           |
       +------------------------------------------------------+

In article <46jkou$6an@panix.com>, Harvey Fishman <fishman@panix.com> wrote:
>
>I have had an ST-15150N in constant operation since April with no
>problems.  I have a friend with very similar experiences.  My ISP has a
>whole passel of them in constant operation for a much longer period of
>time without inordinate failure rates.  The important thing here is to be
>meticulous about the cooling.  The major problem is that the people who
>are too lazy to read and follow the manufacturer's instructions are
>usually the ones with the biggest and baddest mouths.  But I guess that
>that is a consistant pattern of arrogance.
>
>But seriously, if you are unwilling to devote the time and $$$ to do the
>proper cooling then definitely avoid Barracudas.  But if you give them the
>proper care, feeding, and love they will serve you long, well, and fast.

But should one need to go to such extremes with such an expensive
drive?  I don't think so.  We've gone through 5 of the 15150N's in
the last 5 months on one of our news machines at the office -- and
in each case Seagate was at least good about getting a replacement
our via next day air -- but the hassles of bring the machine down
gets one depressed...

At home, I've installed an Atlas.  It appears to be slightly
faster than the 15150N -- I show average thoughput using hdparm
at 9mb..

-- 
Larry D Snyder, Orlando, Florida -- larry@RN.com

In article <46s0so$1uh@rn.com>, Larry D Snyder <larry@rn.com> wrote:
>>But seriously, if you are unwilling to devote the time and $$$ to do the
>>proper cooling then definitely avoid Barracudas.  But if you give them the
>>proper care, feeding, and love they will serve you long, well, and fast.
>
>But should one need to go to such extremes with such an expensive
>drive?  I don't think so.

A Ferrari is an expensive car, but it has much higher maintenance
demands than a Toyota.  The ST15150 was designed to provide bleeding
edge performance (for its time); ease of installation was never one of
its goals and it was sold for servers, not desktop machines.  It is
more like a Ferrari than a Toyota.  It's expensive and you have to
pamper it with forced air cooling on both the top and bottom.  If you
don't want to do that, you bought the wrong drive.

Seagate's 4 GB "desktop" drive is the Hawk 4 (ST15230), which is
slightly slower and cheaper, and runs cooler because of its lower
spindle speed of 5400 rpm instead of 7200.  I sold my ST15150
recently because I didn't want to deal with cooling it, and I'll
probably get an ST15230 soon.

>We've gone through 5 of the 15150N's in
>the last 5 months on one of our news machines at the office -- and
>in each case Seagate was at least good about getting a replacement
>our via next day air -- but the hassles of bring the machine down
>gets one depressed...

Were you cooling them properly?

In article <phrDH6ouJ.4Dz@netcom.com>, Paul Rubin <phr@netcom.com> wrote:
>In article <46s0so$1uh@rn.com>, Larry D Snyder <larry@rn.com> wrote:
>>We've gone through 5 of the 15150N's in
>>the last 5 months on one of our news machines at the office -- and
>>in each case Seagate was at least good about getting a replacement
>>our via next day air -- but the hassles of bring the machine down
>>gets one depressed...
>
>Were you cooling them properly?

Yes, I had a 5" boxer fan blowing over the drive constantly
keeping the drive very cool to touch --


-- 
Larry D Snyder, Orlando, Florida -- larry@RN.com

In article <470pc4$b8k@panix.com>, fishman@panix.com (Harvey Fishman) writes:
> In <phrDH6ouJ.4Dz@netcom.com> phr@netcom.com (Paul Rubin) says
> 
>> Seagate's 4 GB "desktop" drive is the Hawk 4 (ST15230), which is
>> slightly slower and cheaper, and runs cooler because of its lower
>> spindle speed of 5400 rpm instead of 7200.  I sold my ST15150
>> recently because I didn't want to deal with cooling it, and I'll
>> probably get an ST15230 soon.
> 
> The capability that generally causes a drive to use lots of power and thus
> run hot is NOT spindle speed.  It is track-to-track seek.  It takes POWER
> to accelerate the mass of the head assembly from rest and to decelerate it
> back to rest and do both FAST.  Keeping something running at almost any
> constant speed is trivial.

	May I point out something obvious? Such drives run hot even when
there is no access to them for very long times. Guess where heat comes from?
Yeah, spindle motor plus electronics. The tighter the mechanical tolerance
on the spindle the snugger the fit and the more drive power needed to turn
it. And the motor electronics is heftier and draws more standing current.
	This assertion is trivial to verify.
	For those who prefer to "read more about it" then may I recommend
the tiny book for your drive. Here's an example for a Seagate Hawk 2 (2GB,
5400 RPM): Power 10W idle, 12W operating. Model number ST12400N.
	Joe D.

> Harvey
> 
> --
> ----------------------------------------------------------------------------
>  Harvey Fishman   |   It is easy to live by the laws that you agree with.
> fishman@panix.com |          But being law-abiding means obeying
>   718-258-7276    |        even the ones that you think are wrong.
>                                                                                         

In <phrDH6ouJ.4Dz@netcom.com> phr@netcom.com (Paul Rubin) says

> Seagate's 4 GB "desktop" drive is the Hawk 4 (ST15230), which is
> slightly slower and cheaper, and runs cooler because of its lower
> spindle speed of 5400 rpm instead of 7200.  I sold my ST15150
> recently because I didn't want to deal with cooling it, and I'll
> probably get an ST15230 soon.

The capability that generally causes a drive to use lots of power and thus
run hot is NOT spindle speed.  It is track-to-track seek.  It takes POWER
to accelerate the mass of the head assembly from rest and to decelerate it
back to rest and do both FAST.  Keeping something running at almost any
constant speed is trivial.  Drives that do fast track-to-track seeks do
tend to have higher spindle speeds, but that is more-or-less a given.  It
is the seek that has a bigger effect on performance in most cases, and the
user of power in all cases.  However, large numbers (spindle speed) stand
out, while small numbers (seek) sink into the paper.  (And that is why
MTBF is used to describe the real property which is failure rate.)

Also note that this is track-to-track seek, which is really the difficult
property to improve.  Fast average (one third stroke) seek is generally
MUCH easier to lower than track-to-track.  So naturally the average seek
is the thing that appears in most advertisements.

Harvey

--
----------------------------------------------------------------------------
 Harvey Fishman   |   It is easy to live by the laws that you agree with.
fishman@panix.com |          But being law-abiding means obeying
  718-258-7276    |        even the ones that you think are wrong.
                                                                                        

In <46vm9r$b1t@rn.com> larry@rn.com (Larry D Snyder) says

> In article <phrDH6ouJ.4Dz@netcom.com>, Paul Rubin <phr@netcom.com> wrote:
> >In article <46s0so$1uh@rn.com>, Larry D Snyder <larry@rn.com> wrote:
> >>We've gone through 5 of the 15150N's in
> >>the last 5 months on one of our news machines at the office -- and
> >>in each case Seagate was at least good about getting a replacement
> >>our via next day air -- but the hassles of bring the machine down
> >>gets one depressed...
> >
> >Were you cooling them properly?
> 
> Yes, I had a 5" boxer fan blowing over the drive constantly
> keeping the drive very cool to touch --

All too often it is not the quantity of air flow that is important, but
the placement.  In the instance of the ST-15150N the critical components
that MUST be kept cool appear to be the integrated circuits located on the
vertical flexible circuit above the SCSI connector and the large chips on
the top circuit substrate above this.  From what I have been able to
gather, the cooling the HDA itself is a bit less critical.  For one thing
it is a LOT easier to cool, because it is large and relatively even in
temperature.  So air flow almost anywhere will suffice.  But the indicated
chips are concentrated dissipators and need specific direction of air flow
across them at enough speed to cause turbulence.

Harvey

--
----------------------------------------------------------------------------
 Harvey Fishman   |   It is easy to live by the laws that you agree with.
fishman@panix.com |          But being law-abiding means obeying
  718-258-7276    |        even the ones that you think are wrong.
                                                                                                                          

In article <46hcd8$t18@elaine16.Stanford.EDU>,
Earl Mitchell <kla@leland.Stanford.EDU> wrote:
>I'm looking for good reliable 4GB SCSI drive for my PC.
>What should I go for the Seagate Barracuda or Quantum Grand PRix?
>The Quantums appear significantly cheaper but are they better

About 6 months ago I got a Quantum 2GB Empire drive and we've been
quite unhappy with it ever since.  It was cheap at the time, but I
wished many times I had gone with the another drive.  It's failed
once and had to be sent back (took enough diagnostic time to pay for
the drive).  The replacement isn't any more solid seeming, but it does
have a 5 year warranty.

We've just gotten a Micropolis replacement in for it, and I'm quite
happy with it.  I've got a year old 1.7GB Micropolis in the main
computer which has served us well.  Also, Micropolis has an "Advanced
replacement" policy where they'll advance you a drive as soon as you
call to report a failure.  In our case it took 6 days to get the
replacement drive once we had identified the problem (with the Quantum)
because they will NOT advance a drive.

The Fujitsu 4GB drive looks pretty good at $990.  I've had good luck with
2 520MB Fujitsu drives and a 1.2GB Fujitsu, all of which are still working
(4 years later).  The 1.2GB was a real noisy 5"FH mother though!  My
ISP just got one in and it's REAL quiet.

Sean
-- 
"We just wanted to give the band a little more thrust than most other bands."
        - Donald Fagen's reply to why they chose the band name 'Steely Dan'
Sean Reifschneider, Inimitably Superfluous <jafo@tummy.com>
URL: <http://www.tummy.com/xvscan>  XVScan -- HP-UX/Linux X11 scanning software.

jrd@cc.usu.edu (Joe Doupnik) wrote:
>In article <470pc4$b8k@panix.com>, fishman@panix.com (Harvey Fishman) writes:
>> In <phrDH6ouJ.4Dz@netcom.com> phr@netcom.com (Paul Rubin) says
>> 
>>> Seagate's 4 GB "desktop" drive is the Hawk 4 (ST15230), which is
>>> slightly slower and cheaper, and runs cooler because of its lower
>>> spindle speed of 5400 rpm instead of 7200.  I sold my ST15150
>>> recently because I didn't want to deal with cooling it, and I'll
>>> probably get an ST15230 soon.
>> 
>> The capability that generally causes a drive to use lots of power and thus
>> run hot is NOT spindle speed.  It is track-to-track seek.  It takes POWER
>> to accelerate the mass of the head assembly from rest and to decelerate it
>> back to rest and do both FAST.  Keeping something running at almost any
>> constant speed is trivial.
>
>	May I point out something obvious? Such drives run hot even when


Onother consideration with these large drives, unless they are on a server, is that 
they are very loud. I had the 4GB Quantum running under linux for a few weeks. I although
it certainly runs hotter than a normal drive, my main concern was the high pitched wine
it generated. In the end I had to trade it in for two smaller seagates (1GB hawks) which 
are a little quieter.  It was almost impossible to do any work in the same room, at least 
that was my impression. 
-- 

Richard Piper          Email: rpiper@julian.uwo.ca
                              
                       Phone: Work 519-6858300 Ext 8991
                              Page:   519-6453497

fishman@panix.com (Harvey Fishman) wrote:
| The capability that generally causes a drive to use lots of power and thus
| run hot is NOT spindle speed.  It is track-to-track seek.  It takes POWER
| to accelerate the mass of the head assembly from rest and to decelerate it
| back to rest and do both FAST.  

Do you have any evidence that seeking generates more heat than spinning?
Your reasoning seems sound, but my informal "finger testing" contradicts
it.  All the disk drives I have are the same temperature whether they're
idle or in heavy use, and among contemporaries the faster the spindle rate
the higher the temperature.   (The older 5" 3600 RPM drives are hotter,
but that's probably because the platters are larger and heavier, and put
more weight on the spindle bearings.)

---
chris

In article <omlet.309a043fwn@news1.wolfe.net> proverbs@wolfenet.com (Chris Pirih) writes:

  fishman@panix.com (Harvey Fishman) wrote:
  | The capability that generally causes a drive to use lots of power and thus
  | run hot is NOT spindle speed.  It is track-to-track seek.  It takes POWER
  | to accelerate the mass of the head assembly from rest and to decelerate it
  | back to rest and do both FAST.  

  Do you have any evidence that seeking generates more heat than spinning?
  Your reasoning seems sound, but my informal "finger testing" contradicts
  it.  All the disk drives I have are the same temperature whether they're
  idle or in heavy use, and among contemporaries the faster the spindle rate
  the higher the temperature.   (The older 5" 3600 RPM drives are hotter,
  but that's probably because the platters are larger and heavier, and put
  more weight on the spindle bearings.)

The published specifications for most drives show a higher power
consumption while seeking than while just spinning or doing read/write
operations.  Higher power means the drive has to dissipate the energy as
heat.  Still, 7200rpm drives generally have higher power consumption than
5400rpm drives, largely due to energy dissipation through friction, I
expect.

		Leonard

proverbs@wolfenet.com (Chris Pirih) writes:

>Do you have any evidence that seeking generates more heat than spinning?
>Your reasoning seems sound, but my informal "finger testing" contradicts
>it. 

Try touching (use the topside of your fingers if you're brave!) the VCM 
drivers on any high performance disk drive before and after running a 
test that causes extensive seeking for some empirical evidence.

You may windup having a logo branded onto your finger. I wear the Hitachi 
"H" proudly!

-- 
------------------------------------------------------------------------
 Kevin Martinez                 |           Fry's Electronics: Where
 lps@rahul.net                  |         Incompetence is the Standard!
------------------------------------------------------------------------

In article <47dlo5$6f@kelewan.dandelion.com>
           lnz@dandelion.com "Leonard N. Zubkoff" writes:

>The published specifications for most drives show a higher power
>consumption while seeking than while just spinning or doing read/write
>operations.  Higher power means the drive has to dissipate the energy as
>heat.  Still, 7200rpm drives generally have higher power consumption than
>5400rpm drives, largely due to energy dissipation through friction, I
>expect.

That is certainly an issue in a fileserver but the proportion of time in
an ordinary user PC that a drive spends seeking is minimal.

-- 
-----------------------------------------
Lawrence Kirby | fred@genesis.demon.co.uk
Wilts, England | 70734.126@compuserve.com
-----------------------------------------

Kevin Martinez (lps@rahul.net) wrote:
: proverbs@wolfenet.com (Chris Pirih) writes:
: 
: >Do you have any evidence that seeking generates more heat than spinning?
: >Your reasoning seems sound, but my informal "finger testing" contradicts
: >it. 
: 
: Try touching (use the topside of your fingers if you're brave!) the VCM 
: drivers on any high performance disk drive before and after running a 
: test that causes extensive seeking for some empirical evidence.

	Fine, but do you think that those driver IC's generate 
appreciable heat *COMPARED* to the air and bearing friction of the
platters?  I still don't think so.

-- 
 Craig A. Lemon    VE3XCL (Advanced) | http://sunee.uwaterloo.ca/~calemon
 3B UWaterloo Electrical Engineering | calemon@uplink.on.ca (home)
 President, UW Amateur Radio Club    | calemon@sunee.uwaterloo.ca (school)
 Chair, UW IEEE Computer Section     | AX.25: ve3xcl@ve3uow.#swon.on.can.na

tls@cloud9.net (Thor Lancelot Simon) wrote:

>The Atlases I have seem pretty quiet, and reasonably cool.  The Ultrastar 
>DFHSes I have are quiet, but run rather hot.  I don't know about Fujitsu 
>drives in that regard; I have a few 32550N 2GB Barracudas which run 
>(modulo being in rackmount cases with huge huge fans on both sides) very 
>cool, and are quiet enough that at least over the fan noise (not really 
>so bad, with the case in question) I can't hear them seek.  For some 
>reason, though, I always get lousy write performance with the Barracudas; 
>I've noticed this several times.  Does anyone know if there's a mode page 
>bit that might be mis-set for many applications by default?


Yes.  Many SCSI drives come with internal write-caching turned off.
This probably won't dramatically improve long sustained writes of very
large files, but it can make a dramatic difference for most every
other write situation.

The problem is finding a program to change the mode-select parameter
that controls this.  If you have an Adaptec 154x or 174x compatible
card, Roy Neese's SCSICNTL might work.  I think many Unix OS's have
built-in ability to send SCSI commands directly.  But the best I've
found is Matt Thul's utility MSDU, that is if it's a publicly
available product by now.  (Actually I believe even Adaptec's EZ-SCSI
now comes with a utility that only toggles write-caching and does
nothing else)


Phil
pjk@aimnet.com
pjk@netcom.com

[Stuff Deleted]

: The problem is finding a program to change the mode-select parameter
: that controls this.  If you have an Adaptec 154x or 174x compatible
: card, Roy Neese's SCSICNTL might work.  I think many Unix OS's have
: built-in ability to send SCSI commands directly.  But the best I've
: found is Matt Thul's utility MSDU, that is if it's a publicly
: available product by now.  (Actually I believe even Adaptec's EZ-SCSI
: now comes with a utility that only toggles write-caching and does
: nothing else)


Know where I might be able to find these utils?  I'd like to see if the
write-caches on my drives are on.  Related to this heat/noise issue...
I got a Quantum GrandPrix a few months ago and have had no trouble with it,
other than it's high-pitched whine, which I don't seem to notice that much
anymore.  I'm using it 24/7 in my Micron tower case.  I don't believe
there is any _direct_ airflow, but there is nothing on top of it, and
directly below it is a 3.5" floppy, which I believe runs fairly cool.  :)
There are three fans placed in various spots on the tower, but as I said,
the drive doesn't get _direct_ airflow.  Should I be worried?  Most of
the time it just idles.  Anyone know where I can mail-order a fan with
one of those internal power connectors?  

Also, I'm getting a DAT drive in a month or so.  Since my investment in
SCSI will be so large at that point (1 CD-Rom, 2 HDs, 1 DAT), I'm thinking
of taking the plunge and getting some of those super-high-quality granite
digital stuff.  I'll most likely get an active terminator to replace the
termination on my SCSI controller because the CD-ROM doesn't have the
ability to disable the termination on it's end.  This is what makes me
worry about the termination, because a device a backwards as the CD-ROM
with no way to disable termination and and upside-down 50 pin connector
probably has cheezy termination.  Here are my questions...  Will a good
terminator on one end make up for the fact that the terminator on the
other end might not be doing its job up to spec?  Should I spend the extra
$$$ and go for the teflon internal ribbon cable?  I've already got what
I consider to be a perfectly fine ribbon cable.  My feeling is that if
I am encountering any hidden SCSI problems (occasional errors, low xfer
rate), they will go away with this new terminator.  But I just don't know...
Any ideas on this?  Thanks!

David Ross
snakey@cs.umd.edu

the time it just idles.  I


: Phil
: pjk@aimnet.com
: pjk@netcom.com