Why 32768 ?
Hubert Mak
Newbie question for quartz watch frequency:
(1) some say 32768 Hz., some say 32768 KHz., which one is correct ?
(2) Why 32768 ?
Thanks in advance for any info.
Andy Turk
The reason that's significant is that it's easy to build a digital circuit
that
will divide 32,768 beats per second down to 1 beat per second.
32768 Hz is correct. If you want to use kilo-hertz, then you'd write the
figure
as 32.768 KHz (note the decimal).
Andy
Hubert Mak wrote in message <6sk3ek$r...@ds2.acs.ucalgary.ca>...
John Quinn
Curie
discovered its electrical propeties , viz :
A quartz crystal will resonate (vibrate) if mechanical pressure or an
electrical
current is applied to it.
Quartz can be made to resonate at any frequency depending on its shape and
cut.
Most watches use quartz cut and shaped (like a tuning fork) to resonate at
32768 cycles per second (hertz).
To understand the reason for this , you need look no further than the
computer industry :
A BYTE (8 bits ) can contain 2exponent0 to 2Ex8 (0 to 256)
2 BYTES processed simultaneuosly can contain 0 to 32768.(2 to the power of
16)
Computers that process 2 bytes of data concurrently are called 16 bit (8x2)
processors and these are
used in quartz watches. Thus :
For every 32768 vibrations the processor register fills up , the I.C. in the
watch is told a second has
passed (this is passed onto the display) and the registers are reset to zero
to begin again.
This makes them very accurate and of course given the right processors you
would be able to
display time up to 1/32768 of a second , making them excellent devices for
timing athletic events
etc.
compare this to the first mechanical watches which oscillated the balance
wheel at 15-2000 times an hour
and you have some idea of their accuracy.
A very interesting book : 'LONGITUDE' by an author called Dava Sobel gives
an account of the development
of watch mechanisms from the 18th century and why it was vital to get them
accurate ASAP so that the
trade ships of the time could navigate properly.
William L. Bahn
Hubert Mak wrote in message <6sk3ek$r...@ds2.acs.ucalgary.ca>...
>
>Newbie question for quartz watch frequency:
>
>(1) some say 32768 Hz., some say 32768 KHz., which one is correct ?
>(2) Why 32768 ?
>
>Thanks in advance for any info.
Look carefully at the numbers and you should find that they are:
32768 Hz and 32.768 kHz - which are the same thing.
For a watch crystal, you want to be able to use simple clock dividers and
get to a 1s timebase.
32768 Hz is 2^15 Hz. So a 15 stage binary counter will cycle at 1 Hz.
Sherry Hidebrand
William L. Bahn
John Quinn wrote in message <6sk8kd$iib$1...@news.indigo.ie>...
>Quartz is a piezoelectric material occuring naturally in the earth. Pierre
>Curie
>discovered its electrical propeties , viz :
>
>A quartz crystal will resonate (vibrate) if mechanical pressure or an
>electrical
>current is applied to it.
>
>Quartz can be made to resonate at any frequency depending on its shape and
>cut.
>
>Most watches use quartz cut and shaped (like a tuning fork) to resonate at
>32768 cycles per second (hertz).
>
>To understand the reason for this , you need look no further than the
>computer industry :
>
>A BYTE (8 bits ) can contain 2exponent0 to 2Ex8 (0 to 256)
>
Please show the binary representation for 256. Does it fit in a byte?
>2 BYTES processed simultaneuosly can contain 0 to 32768.(2 to the power of
>16)
First, 2^16 is 65536, 32768 is equal to 2^15. A 16 bit (unsigned) binary
number can represent 0 through 65535.
>Computers that process 2 bytes of data concurrently are called 16 bit (8x2)
>processors and these are
>used in quartz watches. Thus :
Hardly. I would be VERY surprised if anyone was foolish enough to put a 16
bit processor in a watch. I would expect them to be 4 bit (if they are even
true micros) but would expect that more likely they are dedicated ASICs
where each point in the data path is no wider than it absolutely has to be.
There is LOTs of time to process data and if you can save the silicon that
an inverter, let along a register, requires you add millions to the bottom
line of the profit sheet over time.
<snip>
>A very interesting book : 'LONGITUDE' by an author called Dava Sobel gives
>an account of the development
>of watch mechanisms from the 18th century and why it was vital to get them
>accurate ASAP so that the
>trade ships of the time could navigate properly.
I will certainly second this endorsement. It is an EXCELLENT book. Fast
reading and filled with a lot of surround that ties so much together - James
Cook, the Bounty, pirates, a bunch of cloak-and-dagger, political intrigue,
international tensions, personal and professional rivalry and sculduggery. A
really fascinating read.
Greg Reid
Where exactly did you get this information? Especially the part about the
quartz crystals shaped like tuning forks?
Where did you receive your education in Digital electronics or Horology
for that matter?
greg
In article <6sk8kd$iib$1...@news.indigo.ie>, "John Quinn"
<quin...@indigo.ie> wrote:
> Quartz is a piezoelectric material occuring naturally in the earth. Pierre
> Curie
> discovered its electrical propeties , viz :
>
> A quartz crystal will resonate (vibrate) if mechanical pressure or an
> electrical
> current is applied to it.
>
> Quartz can be made to resonate at any frequency depending on its shape and
> cut.
>
> Most watches use quartz cut and shaped (like a tuning fork) to resonate at
> 32768 cycles per second (hertz).
>
> To understand the reason for this , you need look no further than the
> computer industry :
>
> A BYTE (8 bits ) can contain 2exponent0 to 2Ex8 (0 to 256)
>
> 2 BYTES processed simultaneuosly can contain 0 to 32768.(2 to the power of
> 16)
>
> Computers that process 2 bytes of data concurrently are called 16 bit (8x2)
> processors and these are
> used in quartz watches. Thus :
>
> For every 32768 vibrations the processor register fills up , the I.C. in the
> watch is told a second has
> passed (this is passed onto the display) and the registers are reset to zero
> to begin again.
>
> This makes them very accurate and of course given the right processors you
> would be able to
> display time up to 1/32768 of a second , making them excellent devices for
> timing athletic events
> etc.
>
> compare this to the first mechanical watches which oscillated the balance
> wheel at 15-2000 times an hour
> and you have some idea of their accuracy.
>
> A very interesting book : 'LONGITUDE' by an author called Dava Sobel gives
> an account of the development
> of watch mechanisms from the 18th century and why it was vital to get them
> accurate ASAP so that the
> trade ships of the time could navigate properly.
>
>
Greg Reid
I was just sent information concerning tuning fork crystals. My apologies
to John, maybe "I" need to go to school.
Greg
In article <rotnappl-030...@nntp.ix.netcom.com>,
Hubert Mak
Thanks to all who responded.
Regards.
Steve Rayner
In the old days they were 1/2" square x & y cut slices held between two
metal plates. Yep, I have an old WW2 Blighley crystal made like this. In
small sizes, it took a high voltage to get proper exitation. For this
reason, the tuning fork type was developed.
Greg Reid (rotn...@ix.netcom.com) wrote:
: greg
: > >
: > >Newbie question for quartz watch frequency:
: > >
: > >(1) some say 32768 Hz., some say 32768 KHz., which one is correct ?
: > >(2) Why 32768 ?
: > >
: > >Thanks in advance for any info.
: > >
: > >
: > >
: > >
--
Steve.
Visit my website at:
http://www.victoria.tc.ca/~ud233/homepage.htm
************* Tu ne cede malis, sed contra audentior ito. ****************
******************************** - Virgil ********************************
******Yield thou not to adversity, but press on the more bravely.**********
Mike McCarty
William L. Bahn <ba...@bfe.com> wrote:
)Please show the binary representation for 256. Does it fit in a byte?
[snip]
100000000
It does not fit in a (regular) byte.
Note that BYTE does *not* mean exactly 8 bits, though that is its
customary meaning. Read Knuth.
--
----
char *p="char *p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
This message made from 100% recycled bits.
I don't speak for DSC. <- They make me say that.
William L. Bahn
Mike McCarty wrote in message <6spapi$5q1$1...@relay1.dsccc.com>...
>In article <6slc11$f6g$1...@newman.pcisys.net>,
>William L. Bahn <ba...@bfe.com> wrote:
>)Please show the binary representation for 256. Does it fit in a byte?
>
>[snip]
>
>100000000
>
>It does not fit in a (regular) byte.
>
>Note that BYTE does *not* mean exactly 8 bits, though that is its
>customary meaning. Read Knuth.
Agreed. But is IS the meaning explicitly defined by the person I was
responding to.
Mike L.
>> 2 BYTES processed simultaneuosly can contain 0 to 32768.(2 to the power of
>> 16)
??? How comes in europe it is always 65536 ;-)? Actually, 32768=2^15.
>> Computers that process 2 bytes of data concurrently are called 16 bit (8x2)
>> processors and these are
>> used in quartz watches. Thus :
Not even swiss army watches use, let alone need, a 16 bit processor. In
fact most watches do not contain any processor at all.
>> For every 32768 vibrations the processor register fills up , the I.C. in the
>> watch is told a second has
>> passed (this is passed onto the display) and the registers are reset to zero
>> to begin again.
Since there is no processor (mostly), this description is far to
complicated.
A toggle flip flop is one of the most basic digital circuits; it changes
its output on every input clock pulse, thus dividing the input frequency
by 2. Now use 15 of these flip flops in a row, the output of one connected
to the input of next, and you divide the crystal frequency of 2^15Hz by
2^15, giving a nice 1Hz clock pulse. Obviously, there is no need for a
processor so far.
>> This makes them very accurate and of course given the right processors you
>> would be able to
>> display time up to 1/32768 of a second , making them excellent devices for
>> timing athletic events
>> etc.
With more sophisticated chips, it would of course be possible to use any
crystal; but the manufacturers seem to prefer crystal frequencies that are
powers of 2, for sake of simplicity. I own a wristwatch, for example, that
contains a 4194304Hz crystal (2^22). But lower frequency crystals normally
lead to less power consumption in digital circuits; that makes 32768Hz
crystals the most common ones in battery powered clocks.
Regards
Michael
Mike McCarty
)>In article <6slc11$f6g$1...@newman.pcisys.net>,
)>William L. Bahn <ba...@bfe.com> wrote:
)>)Please show the binary representation for 256. Does it fit in a byte?
)>
)>[snip]
)>
)>100000000
)>
)>It does not fit in a (regular) byte.
)>
)>Note that BYTE does *not* mean exactly 8 bits, though that is its
)>customary meaning. Read Knuth.
)
)
)Agreed. But is IS the meaning explicitly defined by the person I was
)responding to.
Agreed. I was not pointing a finger at you, just making a general
statement.
Mike
T Hwang
Have anyone heard about Piezo effect?
Tuning fork shaped X-tal, Huh! Some folks looks like forget everything
they learned at school(physics, biology, math, chemistry, etc.) as soon
as they graduate or may be never graduated?
Tony
Steve Rayner wrote:
>
> The small ones are indeed shaped like a tuning fork. Cut one up and look!
> In the old days they were 1/2" square x & y cut slices held between two
> metal plates. Yep, I have an old WW2 Blighley crystal made like this. In
> small sizes, it took a high voltage to get proper exitation. For this
> reason, the tuning fork type was developed.
>
> Greg Reid (rotn...@ix.netcom.com) wrote:
>
> : Where exactly did you get this information? Especially the part about the
> : quartz crystals shaped like tuning forks?
>
> : Where did you receive your education in Digital electronics or Horology
> : for that matter?
>
> : greg
>
> : In article <6sk8kd$iib$1...@news.indigo.ie>, "John Quinn"
> : <quin...@indigo.ie> wrote:
>
> : > Quartz is a piezoelectric material occuring naturally in the earth. Pierre
> : > Curie
> : > discovered its electrical propeties , viz :
> : >
> : > A quartz crystal will resonate (vibrate) if mechanical pressure or an
> : > electrical
> : > current is applied to it.
> : >
> : > Quartz can be made to resonate at any frequency depending on its shape and
> : > cut.
> : >
> : > Most watches use quartz cut and shaped (like a tuning fork) to resonate at
> : > 32768 cycles per second (hertz).
> : >
> : > To understand the reason for this , you need look no further than the
> : > computer industry :
> : >
> : > A BYTE (8 bits ) can contain 2exponent0 to 2Ex8 (0 to 256)
> : >
> : > 2 BYTES processed simultaneuosly can contain 0 to 32768.(2 to the power of
> : > 16)
> : >
> : > Computers that process 2 bytes of data concurrently are called 16 bit (8x2)
> : > processors and these are
> : > used in quartz watches. Thus :
> : >
> : > For every 32768 vibrations the processor register fills up , the I.C. in the
> : > watch is told a second has
> : > passed (this is passed onto the display) and the registers are reset to zero
> : > to begin again.
> : >
> : > This makes them very accurate and of course given the right processors you
> : > would be able to
> : > display time up to 1/32768 of a second , making them excellent devices for
> : > timing athletic events
> : > etc.
Ken
> Hi,
> Have anyone heard about Piezo effect?
> Tuning fork shaped X-tal, Huh! Some folks looks like forget everything
> they learned at school(physics, biology, math, chemistry, etc.) as soon
> as they graduate or may be never graduated?
> Tony
>
My opinion on this subject is probably not going to account for much,
but I too, have heard of and seen tuning fork shaped crystals.
As far as education goes your grammer seems to put you in the latter class.
Ken
Ken
> It appears from your spelling of grammar that this would be true for
> you, too.
>
> Regards
I made no claims to being immune to prosecution. Besides some crystals
used in x-tal's are shaped like tuning forks. Let us please remain on topic.
Ken
Tom MacIntyre
>>My opinion on this subject is probably not going to account for much,
>>but I too, have heard of and seen tuning fork shaped crystals.
>>As far as education goes your grammer seems to put you in the latter =
>class.
>>
>>Ken
>
>It appears from your spelling of grammar that this would be true for
>you, too.
Okay...I have snipped the original tuning fork reference, and I will
follow my interpretation of what the author meant.
>"Most watches use quartz cut and shaped (like a tuning fork) to resonate at
>32768 cycles per second (hertz)."
He didn't mean that the quartz was cut in the shape of a tuning fork.
The parentheses indicate that his meaning was quartz is cut and shaped
to resonate a particular way, not in the same SHAPE as a tuning fork,
but to achieve a similar result.
Am I wrong?
Tom
T Hwang
256 is multiple of 8(1 byte). 8x32=256. 256x128=36728. 8 is in binary
1000, which means it is four bit processor?
Tony
William L. Bahn wrote:
>
> Mike McCarty wrote in message <6spapi$5q1$1...@relay1.dsccc.com>...
> >In article <6slc11$f6g$1...@newman.pcisys.net>,
> >William L. Bahn <ba...@bfe.com> wrote:
> >)Please show the binary representation for 256. Does it fit in a byte?
> >
> >[snip]
> >
> >100000000
> >
> >It does not fit in a (regular) byte.
> >
> >Note that BYTE does *not* mean exactly 8 bits, though that is its
> >customary meaning. Read Knuth.
>
> Agreed. But is IS the meaning explicitly defined by the person I was
> responding to.
T Hwang
Coming from EE background, I did not do well in grammar anyhow. English
is not my mother language either, at least I have an excuse being poor
in grammar and spelling, Ken. Funny, my kids always win English and
French language art trophies at school tho.
Yes, x-tals can be cut at different planes such as x, y, and z cut.
Basically the thickness of element determines the resonanat freq. Some
cuts give more harmonics than others. So depending on application they
choose to how to cut it. Usually in digital electronics, we use high
freq. X-tal(smaller size) PLL circuit as a reference and then divide
it down as necessary. If extreme accuracy is required, we put the x-tal
in an oven to keep it at constant temperature.(commonly called x-tal
oven). Or whole circuitry is hermetically sealed as a single unit.
If it goes bad, you replace the whole thing. There is nothing to repair
or touch.
Tony
T Hwang
It must be in KHz. If it is in Hz, the size of X-tal will be VERY Big.
32768Hz=32.768KHz=VERY big X-tal in size. 32768KHz=32.768MHz more like
it. The idea is use high freq. X-tal for size and divide the freq. down
digitally as much as needed. One ASIC will be able to do it easy as far
as circuitry goes.
William L. Bahn wrote:
>
> >
> >Newbie question for quartz watch frequency:
> >
> >(1) some say 32768 Hz., some say 32768 KHz., which one is correct ?
> >(2) Why 32768 ?
> >
> >Thanks in advance for any info.
>
Hafthor Stefansson
> crystal; but the manufacturers seem to prefer crystal frequencies that
are
> powers of 2, for sake of simplicity. I own a wristwatch, for example,
that
> contains a 4194304Hz crystal (2^22). But lower frequency crystals
normally
> lead to less power consumption in digital circuits; that makes 32768Hz
> crystals the most common ones in battery powered clocks.
A quick note, I'm sure you know this, but forgot to mention this:
Why not just use a 1Hz crystal? Because of limitations of practical
crystals, lower frequencies are generally not found.
The reason for the even power of two frequency is simply because it is much
easier to make a divide by two frequency divider circuit than a divide by
three or five, or some other prime. 32768Hz oscillations, 15 cascaded
dividers later and whammo - 1Hz.
Ken
> > With more sophisticated chips, it would of course be possible to use any
> > crystal; but the manufacturers seem to prefer crystal frequencies that
> are
> > powers of 2, for sake of simplicity. I own a wristwatch, for example,
> that
> > contains a 4194304Hz crystal (2^22). But lower frequency crystals
> normally
> > lead to less power consumption in digital circuits; that makes 32768Hz
> > crystals the most common ones in battery powered clocks.
>
>
> Why not just use a 1Hz crystal? Because of limitations of practical
> crystals, lower frequencies are generally not found.
Since you brought that up I have a question. I have a circuit I've
been wanting to build for some time that calls out a 500kc x-tal.
Anybody know where to find a quantity of one of this seemingly
non-existant device. I've searched for 3 years to no avail except
going to a x-tal house and having one custom made.
Ken
ze...@magicnet.net
> Hi,
> 256 is multiple of 8(1 byte). 8x32=256. 256x128=36728. 8 is in binary
> 1000, which means it is four bit processor?
> Tony
The first LED Pulsar used a 4 bit CPU, the 4004 to be exact.
--
If only we knew that money is only an idea. There is no scarcity or loss connected to it. Nothing cost anything.
William L. Bahn
T Hwang <28103...@home.com> wrote in message
<35F2CC2B...@home.com>...
>Hi,
>256 is multiple of 8(1 byte). 8x32=256. 256x128=36728. 8 is in binary
>1000, which means it is four bit processor?
>Tony
>
I don't know what you are getting at - especially in the context of this
discussion.
What are you trying to say when you say, "256 is multiple of 8 (1byte)"?
Yes, 256 is a multiple of 8, but so is 16 and so is 512. And what is the (1
byte) is the above statement supposed to indicate? Per the thread's context,
a byte is a collection of eight bits. By itself, it can represent 256
states. But in most uses, it cannot represent 256 because, as an unsigned
integer for example, one of those states is usually assigned the value zero
hence you can represent numbers from 0 through 255.
I really am at a loss to understand the "8 is in binary 1000 which means it
is four bit processor?" I think you are confusing the width of the number
field with the representation of a particular number within that field.
Frankly, I think that everyone that thinks that they can deduce the data
width of a micro based on a something like the fact that 32768 (I assume
that was a type when you said 36728) is 2^15 are making similar mistakes.
To first order, the wider the data bus in a processor the more data I can
process per clock cycle - and with a significantly better than equal
trade-off. If I make the the databus 16 bits instead of 8 bits then I can
generally process for more than twice the data per clock cycle - assuming
the data lends itself to 16-bit processing - because I eliminate a LOT of
processing overhead. But the costs are significantly more than an equal
trade-off as well. Again, to a first order, doubling the bus-width will
require (very roughly) a die that is twice as big in both directions -
that's four times the silicon with the associated yield loss.
But since I probably pick up quite a bit more than four times the processing
per clock cycle, why don't I use wide bit processors for everything? First,
there's the issue of whether the data lends itself to wider bit widths. If
it doesn't, then you save very little at significant expense. But more
importantly, the question of whether I NEED more processing power PER CLOCK
CYCLE. If I have lots of clock cycles available, I can implement 128 bit
floating-point math in a four bit processor. How many clock cycles I have
available depends on how fast I clock the system and how much I have to do
per second. In the case of a watch, I really only have to update a register
once every second and check a few flags every second. That's not much to do
considering that I have over 32000 clock cycles to accomplish it in. I also
have to drive the display (If it's an LCD I want to switch it at 100Hz or
so) but that is best done in hardware. Even if I have to do that in siftware
I still have LOTS of extra clock cycles - so I would opt for a narrower
processor to drive costs down. In fact, if I were designing a run of the
mill watch I would want a much slower clock since power consumption is
proportional to clock speed. I suspect (and I'm only guessing) that a 32768
Hz crystal is the slowest crystal that can be made in a small package,
cheaply with the extra high accuracy over temperature and time that a watch
crystal requires. When you think about it, a 65536 Hz clock would actually
make the IC design simpler since that is a even power of 2^4. But the extra
power probably made the greater complexity associated with the slower speed
worth it.
I would be very surprised if any modern watch actually used anything that
could be considered a "mircoprocessor". The first digital watches probably
did - and if so they were almost certainly 4-bit. But a microprocessor, but
its very nature, is intended for flexible applications - i.e., running user
code. A watch is about as far away from this as you can get - it's operation
are VERY rigidly defined. That screams for an ASIC solution. I would
suspect that the vast majority of watches are strictly State Machines with
no software code at all. That will result in the lowest silicon cost and,
given the volume involved, result in huge profit increases over a
microprocessor based solution.
In fact, I wouldn't be surprised if some outfit didn't come up with a watch
IC that has all of the common features (including things like multiple
time-zone and alarms and chronographs) and that nearly all IC's, from the
cheapest to some of the most expensive, use the same exact IC with different
features brought out. Such a company MIGHT enjoy such enormous economies of
scale that they could offer the ICs to watch OEM's for less than the OEM
could produce their own ICs even if they were simpler and smaller.
William L. Bahn
>Hi,
>It must be in KHz. If it is in Hz, the size of X-tal will be VERY Big.
>32768Hz=32.768KHz=VERY big X-tal in size. 32768KHz=32.768MHz more like
>it. The idea is use high freq. X-tal for size and divide the freq. down
>digitally as much as needed. One ASIC will be able to do it easy as far
>as circuitry goes.
Clearly you are grasping at straws. The ubiquitious watch crystal is a 32768
Hz. Yes, 32+kHz. They are very small and are not only cheaper but of greater
accuracy than most oscillator crystals. Why, because clock applications are
very sensitive to accuracy error and because they are produced in quantities
the far exceed nearly any other frequency crystal.
A 32768 Hz crystal is typically 1/3 the price of even other crystals that
are nearly the same frequency (such as 32kHz) with only 1/10 the tolerance
(200ppm instead of 20ppm).
And, before someone else points it out, they do make a 32.768 MHz crystal.
But consider the power implications and you can see why no one would use it
unless they need a resolution beyond that which the 32.768 kHz crystal can
deliver. For instance, it certainly lends itself to measuring things on a
1ms timescale.
William L. Bahn
>Hi,
>Coming from EE background, I did not do well in grammar anyhow. English
>is not my mother language either, at least I have an excuse being poor
>in grammar and spelling, Ken. Funny, my kids always win English and
>French language art trophies at school tho.
>Yes, x-tals can be cut at different planes such as x, y, and z cut.
>Basically the thickness of element determines the resonanat freq. Some
>cuts give more harmonics than others. So depending on application they
>choose to how to cut it. Usually in digital electronics, we use high
>freq. X-tal(smaller size) PLL circuit as a reference and then divide
>it down as necessary. If extreme accuracy is required, we put the x-tal
>in an oven to keep it at constant temperature.(commonly called x-tal
>oven). Or whole circuitry is hermetically sealed as a single unit.
>If it goes bad, you replace the whole thing. There is nothing to repair
>or touch.
But your background appears to be in "digital electronics where cost and/or
power consumption are not the primary design considerations." The subfield
of "digital electronics where cost and/or power consumption are the
overriding design considerations" adopts VERY different approaches. In even
moderately large-volume endeavors, you quickly reach an economy of scale
where you can have a team of ten well-paid engineers work full time for one
year with the ONLY goal being to reduce the bill of materials by ten cents
or eliminate just a single component from the assembly.
As an aside, your ability to write in my native language is FAR beyond any
attempt I might make to write in your native tongue. Your grammar is far
from perfect (as I know mine is) but, unlike many people without this very
legitimate excuse, the meaning of your sentences are USUALLY discernable
with little extra effort.
William L. Bahn
>You're near impossible, Bill. Binary 1000 is decimal 8 and octal 10
>and also hex 8. Byte is a unit of 8 bits not collection of 8 bits.
"Collection of" vs. "unit of" is a semantics and focus-of-discussion issue -
and I don't think it is the point of confusion here.
>I think your knowledge on digital logic is fragmented. Some are bang on
>right and some are way off wrong. If you can't see 1000 in base 2
>as 8 in base 10, I am wondering. One question, 1111 in base 2, what is
>the number in base 10? It is octal 17 and 15 is the answer to the
>question. In hex it is 1F. Confused more now?
I am QUITE conversant with how numbering systems work, thank you. I didn't
say that I can't see how you get "8 is in binary 1000." I said:
I really am at a loss to understand the "8 is in binary 1000 which
means it is four bit processor?"
How does the fact that 8 is represented in binary by 1000 (again, as I
pointed out in my response, in a conventional unsigned integer
interpretation of a 4-bit value) lead to the conclusion that "it is a four
bit processor"?
>Yes, in digital logic, we treat 0 as number. So 0 thru 255 is actually
>representing 256(1 to 256 as we count). Ever heard about double
>precision instruction sets? And floating point instructions?
And in a signed or floating point interpretation 1000 is NOT 8 (in decimal).
So what's your point?
Wuzzzer
accept what movements our watches have and not question it?
Mark
Wuz...@aol.com
$$$$$$$$$$$$$$$$
T Hwang
and also hex 8. Byte is a unit of 8 bits not collection of 8 bits.
right and some are way off wrong. If you can't see 1000 in base 2
as 8 in base 10, I am wondering. One question, 1111 in base 2, what is
the number in base 10? It is octal 17 and 15 is the answer to the
question. In hex it is 1F. Confused more now?
representing 256(1 to 256 as we count). Ever heard about double
precision instruction sets? And floating point instructions?
modern VLSI chips, I gotta know something.
That's why there are ASIC everywhere nowadays.
Remember, in digital logic it's either 1 or 0. Tri-state logic circuitry
is there but it did not really fan out. And scalar vs. vector(pipeline)
processing. And neural system, AI. Just mind boggling. Handling a tiny
watch based on piece of a quartz is nothing. Anyhow All my watches are
vintage mechanical ones. I own lots of chronographs and military
watches. I have one quartz watch, the radio controlled one as a
reference for all my other pieces.
Regards,
Tony
Hafthor Stefansson
> and also hex 8. Byte is a unit of 8 bits not collection of 8 bits.
> I think your knowledge on digital logic is fragmented. Some are bang on
> right and some are way off wrong. If you can't see 1000 in base 2
> as 8 in base 10, I am wondering. One question, 1111 in base 2, what is
> the number in base 10? It is octal 17 and 15 is the answer to the
> question. In hex it is 1F. Confused more now?
Oops, it just F... an honest mistake I'm sure. As someone who also writes
in a second language, I know how hard it is to concentrate on the message
when you are just struggling to string words together.
> Yes, in digital logic, we treat 0 as number. So 0 thru 255 is actually
> representing 256(1 to 256 as we count). Ever heard about double
> precision instruction sets? And floating point instructions?
> Retired after working on computer systems from vacuum tube days to
> modern VLSI chips, I gotta know something.
> That's why there are ASIC everywhere nowadays.
> Remember, in digital logic it's either 1 or 0. Tri-state logic circuitry
> is there but it did not really fan out.
Okay, tri-state is not an alternative to binary logic. It's not trinary.
Tri-state logic is used all the time in real life digital circuits, the
third state being float. This allows, for example, multiple devices to
communicate on the same bus. Only the talker on the bus is not floated.
> And scalar vs. vector(pipeline) processing.
Not really. Pipelining is superscalar. Vector is another deal altogether.
> And neural system, AI. Just mind boggling. Handling a tiny
> watch based on piece of a quartz is nothing. Anyhow All my watches
> are vintage mechanical ones. I own lots of chronographs and military
> watches. I have one quartz watch, the radio controlled one as a
> reference for all my other pieces.
I'm hoping to visit your country (Taiwan, right?) in the next few months.
-HS
[ Jesus is coming - everybody look busy ]
Gray Frierson Haertig
Nebraska Surplus Sales has quite a variety of crystals. My catalog is
not to hand so I don't know if they have any 500 kHz.
New crystals generally aren't all that expensive. The last crystal I
bought new from International Crystal was $16, albeit at a somewhat
higher frequency than 500 kHz.
Gray
--
Telecommunications Engineering
Gray Frierson Haertig & Assoc.
820 North River Street, Suite 100
Portland, Oregon 97227
503-282-2989
503-282-3181 FAX
g...@haertig.com
T Hwang
Excellent idea. Good enough for me.
As far as I am concerned Quartz watch does not have a soul.
Tony
The Baron
William L. Bahn wrote in message <6t2bu1$mhh$1...@newman.pcisys.net>...><35F4A0E0...@home.com>...
>>You're near impossible, Bill. Binary 1000 is decimal 8 and octal 10
>>and also hex 8. Byte is a unit of 8 bits not collection of 8 bits.
>
issue -
>and I don't think it is the point of confusion here.
>
>>right and some are way off wrong. If you can't see 1000 in base 2
>>as 8 in base 10, I am wondering. One question, 1111 in base 2, what is
>>the number in base 10? It is octal 17 and 15 is the answer to the
>>question. In hex it is 1F. Confused more now?
>
>say that I can't see how you get "8 is in binary 1000." I said:
>
> means it is four bit processor?"
>
>pointed out in my response, in a conventional unsigned integer
>interpretation of a 4-bit value) lead to the conclusion that "it is a four
>bit processor"?
>
>>representing 256(1 to 256 as we count). Ever heard about double
>>precision instruction sets? And floating point instructions?
>
go...@my-dejanews.com
or electronic device (cars and Rolexs included). Only living beings are
supposed to have a soul. Well this is what I was lead to believe.
Where would a watch soul go, after the watch dies, 'Watch heaven'??
In article <35F4C249...@home.com>,
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
go...@my-dejanews.com
In article <199809080325...@ladder01.news.aol.com>,
go...@my-dejanews.com
In article <6t4o5u$qk6$1...@excalibur.flash.net>,
"The Baron" <theb...@flash.net> wrote:
> What the hell are you guys talking about? Just get a Rolex !
> William L. Bahn wrote in message <6t2bu1$mhh$1...@newman.pcisys.net>...
> >
> >T Hwang <28103...@home.com> wrote in message
> ><35F4A0E0...@home.com>...
> >>You're near impossible, Bill. Binary 1000 is decimal 8 and octal 10
> >>and also hex 8. Byte is a unit of 8 bits not collection of 8 bits.
> >
> >"Collection of" vs. "unit of" is a semantics and focus-of-discussion
> issue -
> >and I don't think it is the point of confusion here.
> >
> >>I think your knowledge on digital logic is fragmented. Some are bang on
> >>right and some are way off wrong. If you can't see 1000 in base 2
> >>as 8 in base 10, I am wondering. One question, 1111 in base 2, what is
> >>the number in base 10? It is octal 17 and 15 is the answer to the
> >>question. In hex it is 1F. Confused more now?
> >
> >I am QUITE conversant with how numbering systems work, thank you. I didn't
> >say that I can't see how you get "8 is in binary 1000." I said:
> >
> > I really am at a loss to understand the "8 is in binary 1000 which
> > means it is four bit processor?"
> >
> >How does the fact that 8 is represented in binary by 1000 (again, as I
> >pointed out in my response, in a conventional unsigned integer
> >interpretation of a 4-bit value) lead to the conclusion that "it is a four
> >bit processor"?
> >
> >>Yes, in digital logic, we treat 0 as number. So 0 thru 255 is actually
> >>representing 256(1 to 256 as we count). Ever heard about double
> >>precision instruction sets? And floating point instructions?
> >
> >And in a signed or floating point interpretation 1000 is NOT 8 (in
> decimal).
> >So what's your point?
> >
> >
> >
> >
>
>
-----== Posted via Deja News, The Leader in Internet Discussion ==-----