Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.mac.system > #58798 > unrolled thread

[NEWS} Be a Mac OS X Beta tester

Back to article view | Back to comp.sys.mac.system

Contents

  [NEWS} Be a Mac OS X Beta tester Your Name <YourName@YourISP.com> - 2014-04-23 09:14 +1200
    Re: [NEWS} Be a Mac OS X Beta tester billy@MIX.COM - 2014-04-22 22:44 +0000
      Re: [NEWS} Be a Mac OS X Beta tester Alan Browne <alan.browne@FreelunchVideotron.ca> - 2014-04-22 19:17 -0400
        Re: [NEWS} Be a Mac OS X Beta tester billy@MIX.COM - 2014-04-22 23:21 +0000
      Re: [NEWS} Be a Mac OS X Beta tester Your Name <YourName@YourISP.com> - 2014-04-23 19:17 +1200
        Re: [NEWS} Be a Mac OS X Beta tester Lewis <g.kreme@gmail.com.dontsendmecopies> - 2014-04-23 12:49 +0000
          Re: [NEWS} Be a Mac OS X Beta tester billy@MIX.COM - 2014-04-23 14:50 +0000
            Re: [NEWS} Be a Mac OS X Beta tester Your Name <YourName@YourISP.com> - 2014-04-24 09:22 +1200
              Re: [NEWS} Be a Mac OS X Beta tester billy@MIX.COM - 2014-04-24 02:22 +0000
                Re: [NEWS} Be a Mac OS X Beta tester Your Name <YourName@YourISP.com> - 2014-04-24 16:37 +1200
                  Re: [NEWS} Be a Mac OS X Beta tester Invid Fan <invid@loclanet.com> - 2014-04-24 13:19 -0400
                    Re: [NEWS} Be a Mac OS X Beta tester Your Name <YourName@YourISP.com> - 2014-04-25 09:17 +1200
    Re: [NEWS} Be a Mac OS X Beta tester Kevin McMurtrie <mcmurtrie@pixelmemory.us> - 2014-04-23 21:07 -0700
      Re: [NEWS} Be a Mac OS X Beta tester Leonard Blaisdell <leoblaisdell@sbcglobal.net> - 2014-04-23 23:41 -0700
        Re: [NEWS} Be a Mac OS X Beta tester Your Name <YourName@YourISP.com> - 2014-04-24 19:13 +1200
          Re: [NEWS} Be a Mac OS X Beta tester Lewis <g.kreme@gmail.com.dontsendmecopies> - 2014-04-24 20:55 +0000
            Re: [NEWS} Be a Mac OS X Beta tester nospam <nospam@nospam.invalid> - 2014-04-24 17:09 -0400
              Re: [NEWS} Be a Mac OS X Beta tester android <here@there.was> - 2014-04-24 23:19 +0200
            Re: [NEWS} Be a Mac OS X Beta tester Michelle Steiner <michelle@michelle.org> - 2014-04-24 21:03 -0700
            Re: [NEWS} Be a Mac OS X Beta tester Paul Sture <nospam@sture.ch> - 2014-05-10 19:20 +0200
              Re: [NEWS} Be a Mac OS X Beta tester Your Name <YourName@YourISP.com> - 2014-05-11 10:23 +1200
                Re: [NEWS} Be a Mac OS X Beta tester Lewis <g.kreme@gmail.com.dontsendmecopies> - 2014-05-11 15:36 +0000
        Re: [NEWS} Be a Mac OS X Beta tester Kevin McMurtrie <mcmurtrie@pixelmemory.us> - 2014-04-24 21:11 -0700
      Re: [NEWS} Be a Mac OS X Beta tester billy@MIX.COM - 2014-04-24 14:39 +0000
        Re: [NEWS} Be a Mac OS X Beta tester J Burns <burns4@nowhere.com> - 2014-04-25 01:40 -0400
          Re: [NEWS} Be a Mac OS X Beta tester billy@MIX.COM - 2014-04-25 14:15 +0000
            Re: [NEWS} Be a Mac OS X Beta tester J Burns <burns4@nowhere.com> - 2014-04-25 16:07 -0400
              Re: [NEWS} Be a Mac OS X Beta tester Lewis <g.kreme@gmail.com.dontsendmecopies> - 2014-04-27 21:40 +0000
                Re: [NEWS} Be a Mac OS X Beta tester J Burns <burns4@nowhere.com> - 2014-04-27 20:46 -0400
              Re: [NEWS} Be a Mac OS X Beta tester billy@MIX.COM - 2014-04-28 18:27 +0000
      Re: [NEWS} Be a Mac OS X Beta tester J Burns <burns4@nowhere.com> - 2014-04-24 23:02 -0400

Page 2 of 2 — ← Prev page 1 [2]

#59267

In article <vn324b-cp02.ln1@news.chingola.ch>, Paul Sture
<nospam@sture.ch> wrote:
> On 2014-04-24, Lewis <g.kreme@gmail.com.dontsendmecopies> wrote:
> > In message <240420141913450733%YourName@YourISP.com> 
> >
> >> Mac OS X 10.1 and the version of Firefox I was using were fine, but
> >> 10.2 and the newer version are a bit buggy.
> >
> > Really, you're seriously claiming that 10.1 was better than the
> > following versions? Congrats, that's the dumbest thing I've heard today.
> 
> Correct.  10.1 was so bad I was on the point of taking the machine back
> to the shop and asking for my money back. 10.2 was a vast improvement and
> 10.3 was better yet.

10.2.8 is definitely worse than 10.1 on my beige G3 PowerMac. The added
USB card does work better under 10.2, but the computer randomly goes to
sleep, usually at the most inconvenient time, and the only way to get
it back is to force-reboot - that never happened with 10.1 and doesn't
happen if I use Mac OS 9, so it's definitely some sort of bug in 10.2.

[toc] | [prev] | [next] | [standalone]

#59282

In message <110520141023117241%YourName@YourISP.com> 
  Your Name <YourName@YourISP.com> wrote:
> In article <vn324b-cp02.ln1@news.chingola.ch>, Paul Sture
> <nospam@sture.ch> wrote:
>> On 2014-04-24, Lewis <g.kreme@gmail.com.dontsendmecopies> wrote:
>> > In message <240420141913450733%YourName@YourISP.com> 
>> >
>> >> Mac OS X 10.1 and the version of Firefox I was using were fine, but
>> >> 10.2 and the newer version are a bit buggy.
>> >
>> > Really, you're seriously claiming that 10.1 was better than the
>> > following versions? Congrats, that's the dumbest thing I've heard today.
>> 
>> Correct.  10.1 was so bad I was on the point of taking the machine back
>> to the shop and asking for my money back. 10.2 was a vast improvement and
>> 10.3 was better yet.

> 10.2.8 is definitely worse than 10.1 on my beige G3 PowerMac. The added
> USB card does work better under 10.2, but the computer randomly goes to
> sleep, usually at the most inconvenient time, and the only way to get
> it back is to force-reboot - that never happened with 10.1 and doesn't
> happen if I use Mac OS 9, so it's definitely some sort of bug in 10.2.

I ran a beige G3 with 10.2.8 for several years as a server. I had to
reboot it about every 100 days, but other than that, it was rock solid.

It operated as a mailserver as a low-access webserver, and I also used
it for ftp and ssh access. It continued to sit on the network as a spare
machine after I disabled it as a mailserver and when I finally took it
offline in 2007 or so it had an uptime of just under 1 year.

-- 
I hate people who pronounce it "vunerable." And you can't correct them
because they're so vulnerable.

[toc] | [prev] | [next] | [standalone]

#58875

In article <230420142341495471%leoblaisdell@sbcglobal.net>,
 Leonard Blaisdell <leoblaisdell@sbcglobal.net> wrote:

> In article
> <mcmurtrie-AA3673.21072423042014@c-61-68-245-199.per.connect.net.au>,
> Kevin McMurtrie <mcmurtrie@pixelmemory.us> wrote:
> 
> > Should I keep a backup on Time Machine, which constantly corrupts itself 
> > because the machine falls asleep while it's in progress and HFS 
> > journalling doesn't actually work, and can not be repaired by any 
> > utility because it has too much complexity?
> > 
> > (Reformatting yet another Time Machine sparsebundle backup as I write 
> > this...)
> 
> Sigh. You used to be a major proponent for Macs. I delighted in your
> wisdom way back with OS9 and before. Then Apple produced OSX. You
> haven't been content since. No?
> 
> leo

It has always been love/hate with Apple.  I started writing code on an 
Apple ][+ and then later on Macs starting with MacOS 6.  Apple has 
generally been the best about configuration and workflows.  Their 
graphics used to be be top-notch too.  They've always been the worst on 
filesystems and multithreading.  QuickTime, which was very cool when it 
came out, has been abandoned for years and is now officially dead.  As 
for hardware, they alternate between amazingly good and amazingly bad.  
Interoperability has cycled between open and closed.  Trash Can Mac 
begins the bad cycle of hardware and iOS and 10.7 began the closed cycle 
of interoperability.  I don't think I want to wait 2+ years for a good 
cycle.  It's just like the late 1990s when Apple nearly went out of 
business, except there's no Steve Jobs to make a dramatic return.  
There's a good chance that Apple has now gone the way of Microsoft - 
unable to create anything that's not bloated, cumbersome, and existing 
in an isolated ecosystem.  It's a shame because a pure marketing company 
like Google could kill them now.

[toc] | [prev] | [next] | [standalone]

#58850

Kevin McMurtrie <mcmurtrie@pixelmemory.us> writes:

> Should I keep a backup on Time Machine, which constantly corrupts
> itself because the machine falls asleep while it's in progress

My systems, including the spinny disks, never sleep, heh.  I do let
my older displays with fluorescent backlights sleep, but that's it.

> and HFS journalling doesn't actually work, and can not be repaired
> by any utility because it has too much complexity?

It's been fine for me (writing device drivers, and frequently crashing
the machine), but I don't use time machine.  I've only used it once to
restore a friend's system, and luckily, it worked.  I use a journaling
editor, and make copies of work whenever I get worried about losing
it - usually right after something runs as intended.  That and daily
backups have been fine for me.

Billy Y..
-- 
        sub     #'9+1   ,r0             ; convert ascii byte
	add     #9.+1   ,r0             ; to an integer
	bcc     20$                     ; not a number

[toc] | [prev] | [next] | [standalone]

#58880

On 4/24/14, 10:39 AM, billy@MIX.COM wrote:
> Kevin McMurtrie <mcmurtrie@pixelmemory.us> writes:
>
>> Should I keep a backup on Time Machine, which constantly corrupts
>> itself because the machine falls asleep while it's in progress
>
> My systems, including the spinny disks, never sleep, heh.  I do let
> my older displays with fluorescent backlights sleep, but that's it.
>
My 19" fluorescent display uses 24W.  I wonder if fluorescent displays 
are still sold. Some 19" LED displays advertise 18W typical, 25W 
maximum.  I guess on maximum brightness, they would be about the same as 
mine.

Others advertise 14W with Energy Star on.  I wonder if Energy Star runs 
the LEDs at about half brightness.

I experimented with deselecting "Put hard drives to sleep when 
possible." When I let my Mini time out and go to sleep, the wattage 
dropped to 1.2 as before, and I couldn't hear a hum with a stethoscope.

Maybe with the hard drive option selected, the disk could go to sleep 
with the computer awake.  I don't know where to read up on it.

The disk in my Mini uses 0.6W at idle, 0.2 at standby, and 0.1 asleep.

Perhaps Power Nap would prevent a Mac from going to sleep during a Time 
Machine backup.

[toc] | [prev] | [next] | [standalone]

#58889

J Burns <burns4@nowhere.com> writes, quoting me:

> > My systems, including the spinny disks, never sleep, heh.  I do let
> > my older displays with fluorescent backlights sleep, but that's it.
> >
> My 19" fluorescent display uses 24W.  I wonder if fluorescent displays 
> are still sold. Some 19" LED displays advertise 18W typical, 25W 
> maximum.  I guess on maximum brightness, they would be about the same as 
> mine.

I'm not worried about conserving energy (I'm told the electric bill
where I work runs about US$500,000 a month, but I've never actually
seen it), I'm concerned about burnouts (either lamps or inverters)
in the displays.  In the world of TV program production it's quite
common to just leave everything on, 24x7.  Some people think that
helps the gear last longer.

> I experimented with deselecting "Put hard drives to sleep when 
> possible." When I let my Mini time out and go to sleep, the wattage 
> dropped to 1.2 as before, and I couldn't hear a hum with a stethoscope.

I've noticed the spinny drives Apple is curently shipping (Hitachi)
are unusually quiet, and the sound they make when seeking doesn't
scare me (quite the opposite, actually), either.

> Maybe with the hard drive option selected, the disk could go to sleep 
> with the computer awake.

Perhaps, but more likely log file activity would keep it awake.

> I don't know where to read up on it.

Not much here -

| OS X: Saving energy with Sleep
http://support.apple.com/kb/HT5963

And more info here -

| Hard drive sleep
http://www.thexlab.com/faqs/harddrivesleep.html

Billy Y..
-- 
        sub     #'9+1   ,r0             ; convert ascii byte
	add     #9.+1   ,r0             ; to an integer
	bcc     20$                     ; not a number

[toc] | [prev] | [next] | [standalone]

#58897

On 4/25/14, 10:15 AM, billy@MIX.COM wrote:
> J Burns <burns4@nowhere.com> writes, quoting me:
>
>>> My systems, including the spinny disks, never sleep, heh.  I do let
>>> my older displays with fluorescent backlights sleep, but that's it.
>>>
>> My 19" fluorescent display uses 24W.  I wonder if fluorescent displays
>> are still sold. Some 19" LED displays advertise 18W typical, 25W
>> maximum.  I guess on maximum brightness, they would be about the same as
>> mine.
>
> I'm not worried about conserving energy (I'm told the electric bill
> where I work runs about US$500,000 a month, but I've never actually
> seen it), I'm concerned about burnouts (either lamps or inverters)
> in the displays.  In the world of TV program production it's quite
> common to just leave everything on, 24x7.  Some people think that
> helps the gear last longer.

<http://www.quantum.com.pl/baza/pdf/6b4c8ae0f8fe4a4d48404a26c43ce99147390b99.pdf>

Here's a PDF from Portwell.  I don't see a date.  They make industrial 
monitors with LED, CCFL, and FFL backlights.  Their LEDs have a MTBF of 
20,000-50,000 hours, CCFLs 50,000 hours, and FFLs 100,000 hours.

I've read that CCFLs dim with age, so sleeping would preserve brightness.
>
>> I experimented with deselecting "Put hard drives to sleep when
>> possible." When I let my Mini time out and go to sleep, the wattage
>> dropped to 1.2 as before, and I couldn't hear a hum with a stethoscope.
>
> I've noticed the spinny drives Apple is curently shipping (Hitachi)
> are unusually quiet, and the sound they make when seeking doesn't
> scare me (quite the opposite, actually), either.
>

http://www.tomshardware.com/reviews/hdd-reliability-storelab,2681.html
Here's a study of statistics from Storelab, a data-recovery company. 
Compared to market share, the number of Hitachis they received was very 
low, and all had been dropped. The average age was 5 years.

Bad Seagates showed up at a rate 6 times higher.  The average age was 
1.5 years.  The most common problem was a firmware failure.

Western Digitals showed up at twice the rate of Hitachis.  Their average 
age was 1.5 years.  Their heads could overheat at only 45C.  Also, 
pressure on the case could shift the axle, damaging the platters.

Toshiba/Fujitsu appeared at 2.5 times the rate of Hitachi.  Their 
average age was 2 years.  The case could easily be deformed, allowing 
the lubricant to evaporate.

Samsungs also appeared at a rate 2.5 times higher than Hitachi.  The 
most common cause was head crashes.

Axle wedging was a common cause of failure.  The loss of lubricant in 
Toshiba/Fijitsus was one cause.  There's more risk with high-capacity 
drives because the axles carry more weight.

The article doesn't prove one brand is better, but some models have 
problems.  The article suggests to me that perhaps neither run time nor 
sleep/wake cycles have much effect on failure rates.

Years ago, I downloaded Hitachi's 188-page manual for the drive I 
installed in my Mini.  A computer could allow the drive to use several 
stages of readiness.

Immediately after action, it would be at Performance Idle (2W).  It 
could step down to Active Idle (0.8W), with the heads parked over the 
platters and a response time of 20ms.

At Low Power Idle (0.6W), the heads would be unloaded and parked. 
Response time was 300ms.

At Standby (0.2W), the motor would be off. It would take typically 2.5s 
and as long as 9.5s to come to idle.  Standby wasn't sleep.

In a laptop, I imagine parking the heads at every chance could increase 
reliability.

[toc] | [prev] | [next] | [standalone]

#58962

In message <ljef9r$9fg$1@dont-email.me> 
  J Burns <burns4@nowhere.com> wrote:
> I've read that CCFLs dim with age, so sleeping would preserve brightness.

They dim with AGE, not with use, so it really makes no difference.

-- 
Chefet, Chefet, thought Dios. Maker of rings, weaver of metal. Now he's
out of our heads, and see how his nails grow into claws...

[toc] | [prev] | [next] | [standalone]

#58974

On 4/27/14, 5:40 PM, Lewis wrote:
> In message <ljef9r$9fg$1@dont-email.me>
>    J Burns <burns4@nowhere.com> wrote:
>> I've read that CCFLs dim with age, so sleeping would preserve brightness.
>
> They dim with AGE, not with use, so it really makes no difference.
>
Where can I read up on that?  Since the 1960s, the charts I've seen show 
fluorescent aging as hours of use.  The UV gradually breaks down the 
phosphors.  CFLs are supposed to dim faster than traditional tubes. 
They may be down to 80% in only 1,000 hours of use and 60% in 4,000 
hours. I haven't read about CCFLs.

Dirt also causes dimming, and an unused tube could get dirty.

[toc] | [prev] | [next] | [standalone]

#58986

J Burns <burns4@nowhere.com> writes:

> <http://www.quantum.com.pl/baza/pdf/6b4c8ae0f8fe4a4d48404a26c43ce99147390b99.pdf>
> 
> Here's a PDF from Portwell.  I don't see a date.  They make industrial 
> monitors with LED, CCFL, and FFL backlights.  Their LEDs have a MTBF of 
> 20,000-50,000 hours, CCFLs 50,000 hours, and FFLs 100,000 hours.

Interesting reading, thanks.

> I've read that CCFLs dim with age, so sleeping would preserve brightness.

I wish I'd have measured mine when they were new.  I have noticed some
(not much) dimming, and they are now a bit slower to come up to "full"
level.

> http://www.tomshardware.com/reviews/hdd-reliability-storelab,2681.html
> Here's a study of statistics from Storelab, a data-recovery company. 
> Compared to market share, the number of Hitachis they received was very 
> low, and all had been dropped. The average age was 5 years.

Tom's being very careful not to step on anyone's toes, but, regardless
of the small sample size, I think these data are useful.  Particularly
the obserations about <1GB (or, 1GB and up) drives, which match my own 
experience.

> The article doesn't prove one brand is better, but some models have 
> problems.  The article suggests to me that perhaps neither run time nor 
> sleep/wake cycles have much effect on failure rates.

Evidentally not with today's drives, as compared to mechanical problems.

> Years ago, I downloaded Hitachi's 188-page manual for the drive I 
> installed in my Mini.  A computer could allow the drive to use several 
> stages of readiness.
> 
> Immediately after action, it would be at Performance Idle (2W).  It 
> could step down to Active Idle (0.8W), with the heads parked over the 
> platters and a response time of 20ms.
> 
> At Low Power Idle (0.6W), the heads would be unloaded and parked. 
> Response time was 300ms.
> 
> At Standby (0.2W), the motor would be off. It would take typically 2.5s 
> and as long as 9.5s to come to idle.  Standby wasn't sleep.

I haven't looked at what other makers drives can do, but I'm sure this
appealed to Apple.

Billy Y..
-- 
        sub     #'9+1   ,r0             ; convert ascii byte
	add     #9.+1   ,r0             ; to an integer
	bcc     20$                     ; not a number

[toc] | [prev] | [next] | [standalone]

#58870

On 4/24/14, 12:07 AM, Kevin McMurtrie wrote:

> Should I keep a backup on Time Machine, which constantly corrupts itself
> because the machine falls asleep while it's in progress and HFS
> journalling doesn't actually work, and can not be repaired by any
> utility because it has too much complexity?
>

I've read that sometimes a disk with journaling can't be repaired 
because it won't mount.  In that case, one can use Terminal to mount it 
without journaling (-j).

[toc] | [prev] | [standalone]

Page 2 of 2 — ← Prev page 1 [2]

Back to top | Article view | comp.sys.mac.system

csiph-web