You touched on a couple of things that I've noticed, as I also work
mostly on Toshiba (followed by Dell, Gateway and Lenovo).

Actually, I had more problems with AC adapter sockets on motherboards
themselves back in the old ("Toshiba built") days. That is the one --
the ONLY -- area where I think things got better (almost all later
Toshibas mount the DC input jack on the case rather than the motherboard
and have a short cable to the motherboard). The sockets still fail,
which is a user error (mechanical stress), but now you can replace the
socket without having to replace the motherboard (or remove and replace
a component soldered to the motherboard).

However, I've seen a HUGE number of motherboard failures in Toshiba
units that I didn't used to see. Apparently (because I don't do this
work myself, but I talk to the people I send motherboards out to) the
motherboards themselves are now so thin and flexible that the BGA
soldering of the chipsets to the motherboard fails. And I see a LOT
more problems with Toshiba units in this regard than with some other
brands (the A105 and A135 were particularly bad ... a huge number of
motherboard failures).

Another big issue -- and one that is pushing me away from Toshiba,
frankly -- is problems caused by overheating. The same guys who replace
BGA chipsets on motherboards tell me that occasionally they can just
"reflow" the chipset and it fixes the BGA connections. BUT, too often,
the chipset itself is fried by overheating. The chipsets have heatsinks
also, in most modern laptops. But when the cooling system fails (e.g.
the fan and heatsink become clogged with dirt, dust and hair), the CPUs
handle it relatively gracefully (they "throttle" themselves and prevent
their own destruction) but the chipsets don't handle it so well and are
often destroyed. Not a huge difference between Toshiba and other brands
in that regard, EXCEPT .... Dell and Lenovo and even Gateway provide a
door (removeable panel) on the bottom of the laptop explicitly for
cleaning the CPU cooling system (heatsink and fan). Toshiba used to do
that in some older models (the A40's, for example), but in the newer
units (again, starting, it seems, with the A100/105 series) they don't.
THERE IS NO WAY TO CLEAN THE CPU HEATSINK AND FAN, other than a total
disassembly of the unit (remove the motherboard from the case, because
the CPU is now on the bottom of the motherboard in most units). Of
course, this never happens, and we have, again, a huge number of
motherboard failures. [When I do disassemble a unit, it's not unusual
to find a couple of CUBIC INCHES of "hairball", often so bad that the
fan blades no longer are even capable of turning.]

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.


me/2 wrote:

>
> I'll have to agree with you there. My last 10 years of employment
> before retiring around 3 years ago was as a senior tech at a Toshiba
> Premier ASP. I was there when Toshiba made the transition from
> notebooks being designed and built by Toshiba to notebooks being
> speced by Toshiba and being designed and built by Compal. This was
> followed by Toshiba using other Taiwanese ODS like Inventec, Quanta
> and ASUStek.
>
> During the Toshbia designed and built days we saw more notebooks in
> for service that had software issues and not failed hardware. Once
> they went over to the ODMs we saw just the opposite. Either outright
> hardware failures like the rear mounted ac adapter jack that was
> soldered directly to the system board and easily broke off damaging
> the system board beyond normal repair or poorly designed cooling
> systems that were all but impossible for the end user to keep clean
> and working correctly. Also virtually all the class action lawsuits
> that Toshiba was hit with in those days were about ODM designed and
> bult models.
>
> me/2

"Barry Watzman" <WatzmanNOSPAM@neo.rr.com> wrote in message
news:hcnkd7$kcu$1@news.eternal-september.org...
> Re: "I am not convinced that there is a decline in built quality either"
>
> If you have been taking apart a variety of models made over a 15-year
> period, your doubts about that would disappear very quickly. They don't
> build them like they used to. It's only partly cost related, the pressure
> to make things smaller and lighter (and, especially, thinner) also
> contributes. The circuitry, the ICs and other components, are not the
> problem. It's the overall mechanical assembly, everything from the
> thickness of the motherboard and the thickness of the foil traces to the
> all-plastic (thin, cheap plastic) cases. The build quality is not what it
> used to be. Again, the problem is not the components (which includes the
> hard drives), it's how those components are assembled.
>

The dimensions that you refer to for the motherboard are a function of the
frequency of the signals that the traces are required to convey. As the
various clock speeds rise, the traces have to become narrower to minimise
the series inductance. At the same time the traces have to be closer to the
ground plane to allow the signals to propagate with the minimum of
reflections. This points to thinner motherboards with narrower traces. I
agree with you about the plastic mouldings though. Breakage is all too
common, but worse is the tendency for the lid to crack around the hinges
precisely because there is unsufficient meat to withstand the flexing as the
lid is opened and closed.

>
>
> BillW50 wrote:
>> In news:hclr0h$thk$1@news.eternal-september.org,
>> Barry Watzman typed on Mon, 02 Nov 2009 00:34:06 -0500:
>>> 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".
>>>
>>> The differences are not components or specifications, but rather what
>>> I call "build quality". The decline in "build quality" comes from
>>> pressure to cut down on cost, weight and size, all of which contribute
>>> to systems that are electrically well designed but which are
>>> mechanically flimsy and of inferior quality compared to the way that
>>> laptops were built 4 to 8 years ago.
>>
>> I don't know... as computers are always going to get more powerful and
>> less expensive over time. And I am not convinced that there is a decline
>> in built quality either. As laptops have also improved over the years
>> too. Like getting rid of florescent lamps and inverters and trading them
>> for a row of bright LEDs. Which are much more reliable and not as
>> sensitive to shock.
>>
>> Speaking about not sensitive to shock, we are only a few years away from
>> SSDs replacing old mechanical hard drives. I too also believe this is far
>> better than the older method. And speaking of mechanical hard drives,
>> these too have been improved over the years. Some even sport anti-shock
>> features. They have also increased in capacity and also dropped in price.
>>
>> Take my old Epson PX-8 built back in 1984. I like to think of it as the
>> first netbook. As it is about the same size and shape. But the display is
>> terrible and it is heavy. Besides it is very limited compared to today's
>> standards. Although it is well built, nonetheless. Although it also cost
>> $2000 (including the RAMDrive and extras) back then. I do love the
>> Epson's keyboard, but for that price you should get a great keyboard.
>> Even then and now though, I would have traded it for one of today's
>> netbooks in a heartbeat. ;-)
>>

The answers have certainly given me plenty to think about. I shall take a
"tinny" cd along to the store and ask to hear it played. If the sound is
not too bad then I think that I shall stick with the Acer.
I have speakers that I use with my present laptop but that is clumsy; they
are powered from mains; I see that some nowadays are powered from a usb
slot.
Thanks again.




"Richard Bonner" <ak621@chebucto.ns.ca> wrote in message
news:hch86p$er5$1@Kil-nws-1.UCIS.Dal.Ca...
> Stewart (him@invalid.supanet.co.uk) wrote:
>> I want to buy a new 15" laptop and one of the main criteria is that it
>> should produce reasonable quality sound when we are listening to the bbc
>> iplayer, dvds etc..
>
> (Snip)
>> The Acer laptops are shown as having dolby sound but I am not sure if
>> this makes a difference. it is not easy to try them out in a store
>> otherwise I would take along a cd and listen to that.
>
> *** I would think that any salesperson would allow customers to try
> out audio if it means a sale.
>
>
>> Does anyone know of a brand that has reasonable sound output? My price
>> range is between £450 and £550.
>
> *** The likelihood today of getting decent sound in anything in that
> price range is low. They make them as cheaply as possible. If you want to
> get a pro model, they might sound better but will cost you considerably
> more.
>
> I think the suggestions here regarding external speakers may end up
> being your best solution as far as sound goes, but would reduce
> portability, and would extend the set-up time. )-:
>
> --
> Richard Bonner
> http://www.chebucto.ca/~ak621/DOS/

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.]


M.I.5¾ wrote:
> "Barry Watzman" <WatzmanNOSPAM@neo.rr.com> wrote in message
> news:hcnkd7$kcu$1@news.eternal-september.org...
>> Re: "I am not convinced that there is a decline in built quality either"
>>
>> If you have been taking apart a variety of models made over a 15-year
>> period, your doubts about that would disappear very quickly. They don't
>> build them like they used to. It's only partly cost related, the pressure
>> to make things smaller and lighter (and, especially, thinner) also
>> contributes. The circuitry, the ICs and other components, are not the
>> problem. It's the overall mechanical assembly, everything from the
>> thickness of the motherboard and the thickness of the foil traces to the
>> all-plastic (thin, cheap plastic) cases. The build quality is not what it
>> used to be. Again, the problem is not the components (which includes the
>> hard drives), it's how those components are assembled.
>>
>
> The dimensions that you refer to for the motherboard are a function of the
> frequency of the signals that the traces are required to convey. As the
> various clock speeds rise, the traces have to become narrower to minimise
> the series inductance. At the same time the traces have to be closer to the
> ground plane to allow the signals to propagate with the minimum of
> reflections. This points to thinner motherboards with narrower traces. I
> agree with you about the plastic mouldings though. Breakage is all too
> common, but worse is the tendency for the lid to crack around the hinges
> precisely because there is unsufficient meat to withstand the flexing as the
> lid is opened and closed.
>
>>
>> BillW50 wrote:
>>> In news:hclr0h$thk$1@news.eternal-september.org,
>>> Barry Watzman typed on Mon, 02 Nov 2009 00:34:06 -0500:
>>>> 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".
>>>>
>>>> The differences are not components or specifications, but rather what
>>>> I call "build quality". The decline in "build quality" comes from
>>>> pressure to cut down on cost, weight and size, all of which contribute
>>>> to systems that are electrically well designed but which are
>>>> mechanically flimsy and of inferior quality compared to the way that
>>>> laptops were built 4 to 8 years ago.
>>> I don't know... as computers are always going to get more powerful and
>>> less expensive over time. And I am not convinced that there is a decline
>>> in built quality either. As laptops have also improved over the years
>>> too. Like getting rid of florescent lamps and inverters and trading them
>>> for a row of bright LEDs. Which are much more reliable and not as
>>> sensitive to shock.
>>>
>>> Speaking about not sensitive to shock, we are only a few years away from
>>> SSDs replacing old mechanical hard drives. I too also believe this is far
>>> better than the older method. And speaking of mechanical hard drives,
>>> these too have been improved over the years. Some even sport anti-shock
>>> features. They have also increased in capacity and also dropped in price.
>>>
>>> Take my old Epson PX-8 built back in 1984. I like to think of it as the
>>> first netbook. As it is about the same size and shape. But the display is
>>> terrible and it is heavy. Besides it is very limited compared to today's
>>> standards. Although it is well built, nonetheless. Although it also cost
>>> $2000 (including the RAMDrive and extras) back then. I do love the
>>> Epson's keyboard, but for that price you should get a great keyboard.
>>> Even then and now though, I would have traded it for one of today's
>>> netbooks in a heartbeat. ;-)
>>>
>
>

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.]

M.I.5¾ wrote:
> "Barry Watzman" <WatzmanNOSPAM@neo.rr.com> wrote in message
> news:hcnkd7$kcu$1@news.eternal-september.org...
>> Re: "I am not convinced that there is a decline in built quality either"
>>
>> If you have been taking apart a variety of models made over a 15-year
>> period, your doubts about that would disappear very quickly. They don't
>> build them like they used to. It's only partly cost related, the pressure
>> to make things smaller and lighter (and, especially, thinner) also
>> contributes. The circuitry, the ICs and other components, are not the
>> problem. It's the overall mechanical assembly, everything from the
>> thickness of the motherboard and the thickness of the foil traces to the
>> all-plastic (thin, cheap plastic) cases. The build quality is not what it
>> used to be. Again, the problem is not the components (which includes the
>> hard drives), it's how those components are assembled.
>>
>
> The dimensions that you refer to for the motherboard are a function of the
> frequency of the signals that the traces are required to convey. As the
> various clock speeds rise, the traces have to become narrower to minimise
> the series inductance. At the same time the traces have to be closer to the
> ground plane to allow the signals to propagate with the minimum of
> reflections. This points to thinner motherboards with narrower traces. I
> agree with you about the plastic mouldings though. Breakage is all too
> common, but worse is the tendency for the lid to crack around the hinges
> precisely because there is unsufficient meat to withstand the flexing as the
> lid is opened and closed.
>

On Mon, 02 Nov 2009 19:54:40 -0500, Barry Watzman
<WatzmanNOSPAM@neo.rr.com> 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

me/2 wrote on Tue, 03 Nov 2009 18:22:39 -0700:
> On Mon, 02 Nov 2009 19:54:40 -0500, Barry Watzman
> <WatzmanNOSPAM@neo.rr.com> 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" <WatzmanNOSPAM@neo.rr.com> wrote in message
news:hcpgof$p5o$1@news.eternal-september.org...
>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" <WatzmanNOSPAM@neo.rr.com> wrote in message
news:hcpgof$p5o$1@news.eternal-september.org...
>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" <WatzmanNOSPAM@neo.rr.com> wrote in message
> news:hcpgof$p5o$1@news.eternal-september.org...
>> 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" <WatzmanNOSPAM@neo.rr.com> wrote in message
news:hcrv6k$pfp$1@news.eternal-september.org...
> 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" <WatzmanNOSPAM@neo.rr.com> wrote in message
>> news:hcpgof$p5o$1@news.eternal-september.org...
>>> 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 (WatzmanNOSPAM@neo.rr.com) 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/