laptop audio

me/2 wrote on Tue, 03 Nov 2009 18:22:39 -0700:
> On Mon, 02 Nov 2009 19:54:40 -0500, Barry Watzman
> <> wrote:
>
> [snipped]
>
> :>What's sad is that this stuff is just plain dumb. There is no reason
> :>for it, it doesn't really save anything, but "that's the way they are".
> :>
> :>For no apparent reason.
>
> Not to sound cynical but to me there is a very apparent reason. Built
> that way they will, in most cases, work fine until the end of the
> usual 1 year warranty period. Once it's no longer the manufacturer's
> problem they would just as soon you buy a new unit than pay to have
> one repaired. For example, that's why in quite a few cases the cost to
> have a display or system board replaced is greater than the original
> cost of the notebook. One thing I did notice in my 10+ years of
> working exclusively on Toshiba is that the "business class" notebooks
> that typically came with a 3 year warranty tended to have much fewer
> hardware issues than the "consumer class" notebooks that typically
> came with a 1 year warranty. Also the business class systems were the
> last to get turned over to the ODMs at which time the standard
> warranty dropped to 1 year on quite a few of them.
>
> BTW, I'm still using a 7 year old Toshiba 5205 that still works as
> good as new except for 1 speaker failing. It has a 2.4ghz mobile P4,
> 1gb RAM and a 1600x1200 WXGA display. Other than having to use the
> standard vga driver it runs the windows 7 rc about as good as it ran
> the original windows xp pro. There is no windows 7 support for some of
> the proprietary stuff like the SD card slot, the bluetooth module, the
> select bay card reader or the select bay tv tuner so eventually the
> factory software will be going back on.
>
> me/2

I have been purchasing laptops since '84. And all except one are doing
just fine (except for the original batteries). So while I believe there
are some makes and models which doesn't last, there are many makes and
models that just keeps on going and shows no signs of ever dying.

It isn't the after warranty failure that seems to get me. But rather it
is the advancing technology which makes them obsolete to me. So I
continue to purchase newer and newer models. I usually get 3 to 6 years
out of a machine before I replace it with newer technology.

It was the 90's which seems to have turned out the most unreliable
hardware IMHO. Some of those in the 80's wasn't too hot either. The
modern day hardware I have, I still see lasting over 25 years from now.
Although I don't think I'll still be using them for everyday use for
that long. But heck, you never know. As I am not that impressed with
Vista and Windows 7. And I probably won't be with Windows 8 and 9
either. ;-)

--
Bill
Asus EEE PC 702G4 ~ 2GB RAM ~ 16GB-SDHC
Xandros Linux (build 2007-10-19 13:03)

Re: "Not to sound cynical but to me there is a very apparent reason.
Built that way they will, in most cases, work fine until the end of the
usual 1 year warranty period"

Having worked for multiple PC mfgrs., I can tell you that manufacturers
do not think that way. It does not make for repeat customers; sure, it
gets the customer to buy a new PC, but if yours failed "too soon", it's
more likely to be replaced with another brand (and it also makes for bad
reliability reviews in "consumers reports" and other publications and
surveys).

The nice thing about PCs, from a manufacturers perspective, is that they
are self-obsoleting, even without actual failure. A 3 year old PC is
obsolete, even if it still works perfectly. Add to that the OS upgrade
cycle (most people will not upgrade a PC from Windows X to Windows X+1,
they will replace the PC). And, finally, a very real reality is that
most people will even replace a PC rather than just reinstall Windows.
All it takes is a bad or corrupted hard drive, a bad virus/malware
infection or even just the routine "registry bloat" in a PC more than 2
to 3 years old, and you have a customer in the market for a replacement
laptop. However, if the previous one had poor hardware reliability ...
the replacement is more likely to be another brand.

As for older PCs, however, I make a small part-time business out of
refurbing and selling 1410/15 and 2410/15 laptops. They are still very
usable, and I get $175 - $225 for them, on E-Bay and sometimes locally.

"Barry Watzman" <> wrote in message
news:hcpgof$p5o$...
>I was referring to the THICKNESS of the copper layer, not the WIDTH of the
>traces (I was also referring to the thickness of the board material itself
>(the fiberglass)). While there is some slight issue with regard to
>frequency, this is primarily a quality issue, and when the layers get very
>thin, they are more prone to being cracked if the circuit board is flexed.
>

The thickness also contributes to the series inductance so as the width
reduces, I guess the thickness has to as well.

> Even so, the dimensions have very little to do with the frequency of the
> signals. Not nothing, but not that much.
>

Not actually the case. Larger conductors means greater series inductance
and that in turn means limits on the propagation of higher frequencies.
This is why higher frequency transformers (found in the larger switch mode
PSUs) are wound with multi stand wire. The series inductance of a solid
single wire is too high for the transformer to function properly. There is
also the skin effect to consider which is a second reason for the use of
stranded wire.

Similarly, higher frequency signals need to be closer to a ground plane for
effective propagation (think 80 conductor hard drive PATA cables here).

> [A good example is Gigabyte modern Core i5 and i7 motherboards, which make
> a specific point of their "3 oz copper" ((weight of copper per unit of
> area) which is quite thick as circuit boards go) and thickness .... and
> these are clearly higher frequency boards than laptops.]
>

Not had a look at one myself, but is the trace narrower than would normally
be the case to compensate?

"Barry Watzman" <> wrote in message
news:hcpgof$p5o$...
>I was referring to the THICKNESS of the copper layer, not the WIDTH of the
>traces (I was also referring to the thickness of the board material itself
>(the fiberglass)). While there is some slight issue with regard to
>frequency, this is primarily a quality issue, and when the layers get very
>thin, they are more prone to being cracked if the circuit board is flexed.
>
> Even so, the dimensions have very little to do with the frequency of the
> signals. Not nothing, but not that much.
>
> [A good example is Gigabyte modern Core i5 and i7 motherboards, which make
> a specific point of their "3 oz copper" ((weight of copper per unit of
> area) which is quite thick as circuit boards go) and thickness .... and
> these are clearly higher frequency boards than laptops.]
>

Bofore hitting return, I was also going to point out that at the higher
centimetric microwave frequencies, the normal PCB substrate is no good at
all, but microwave circuits can be printed onto a ceramic substrate, but the
traces are very thin and narrow (and made of silver). Once millimetric
frequencies are encountered the traces cannot be more than a few atoms
thick.

Digital data signals are not the same as analog signals.

But more to the point, a multi-GHz CPU only has those frequencies within
the CPU. Externally, the signal frequencies are much lower.


M.I.5¾ wrote:
> "Barry Watzman" <> wrote in message
> news:hcpgof$p5o$...
>> I was referring to the THICKNESS of the copper layer, not the WIDTH of the
>> traces (I was also referring to the thickness of the board material itself
>> (the fiberglass)). While there is some slight issue with regard to
>> frequency, this is primarily a quality issue, and when the layers get very
>> thin, they are more prone to being cracked if the circuit board is flexed.
>>
>> Even so, the dimensions have very little to do with the frequency of the
>> signals. Not nothing, but not that much.
>>
>> [A good example is Gigabyte modern Core i5 and i7 motherboards, which make
>> a specific point of their "3 oz copper" ((weight of copper per unit of
>> area) which is quite thick as circuit boards go) and thickness .... and
>> these are clearly higher frequency boards than laptops.]
>>
>
> Bofore hitting return, I was also going to point out that at the higher
> centimetric microwave frequencies, the normal PCB substrate is no good at
> all, but microwave circuits can be printed onto a ceramic substrate, but the
> traces are very thin and narrow (and made of silver). Once millimetric
> frequencies are encountered the traces cannot be more than a few atoms
> thick.
>
>

"Barry Watzman" <> wrote in message
news:hcrv6k$pfp$...
> Digital data signals are not the same as analog signals.
>
> But more to the point, a multi-GHz CPU only has those frequencies within
> the CPU. Externally, the signal frequencies are much lower.
>

True, but they are higher in frequency than they used to be. CPU clock
frequencies would certainly not propagate well along regular motherboard
traces.


>
> M.I.5¾ wrote:
>> "Barry Watzman" <> wrote in message
>> news:hcpgof$p5o$...
>>> I was referring to the THICKNESS of the copper layer, not the WIDTH of
>>> the traces (I was also referring to the thickness of the board material
>>> itself (the fiberglass)). While there is some slight issue with regard
>>> to frequency, this is primarily a quality issue, and when the layers get
>>> very thin, they are more prone to being cracked if the circuit board is
>>> flexed.
>>>
>>> Even so, the dimensions have very little to do with the frequency of the
>>> signals. Not nothing, but not that much.
>>>
>>> [A good example is Gigabyte modern Core i5 and i7 motherboards, which
>>> make a specific point of their "3 oz copper" ((weight of copper per unit
>>> of area) which is quite thick as circuit boards go) and thickness ....
>>> and these are clearly higher frequency boards than laptops.]
>>>
>>
>> Bofore hitting return, I was also going to point out that at the higher
>> centimetric microwave frequencies, the normal PCB substrate is no good at
>> all, but microwave circuits can be printed onto a ceramic substrate, but
>> the traces are very thin and narrow (and made of silver). Once
>> millimetric frequencies are encountered the traces cannot be more than a
>> few atoms thick.
>>

> Richard Bonner wrote:
> > *** I have a couple of friends in, or associated with, the laptop repair
> > business. Their general consensus is that the majority of today's laptops
> > are much more poorly built than older models, with components that are
> > underspecified or are not tested for value accuracy.

Barry Watzman () wrote:
> I have worked for laptop manufacturers, service laptops and teach IT
> (specifically A+ certification and networking courses) at a local college.

> I disagree with the part of your post which stated "components that are
> underspecified or are not tested for value accuracy"; but I do agree
> with "today's laptops are much more poorly built than older models".
(Snip)

*** I mispoke there. I should have typed "just specified". That is,
the components just meet the requirements or have less headroom than they
once did.

As for the "not tested", I meant at the laptop end. Components are
tested and graded at the manufacturing plant, but it costs the laptop
manufacturers time and money to retest them. So I believe fewer do it,
instead trusting the manufacturers to properly grade them, and that the
components have not changed value between making and installing.

--
Richard Bonner
http://www.chebucto.ca/~ak621/DOS/