Sign in with
Sign up | Sign in
Your question

Computers can still count, can't they?

Last response: in Home Theatre
Share
Anonymous
March 9, 2005 3:44:00 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

Computers are supposed to be pretty good at counting things.

So why does my Samsung SIR-T451 digital STB, when I ask it
to tell me the signal strength of a particular channel, present
me with a picture of a group of closely spaced line segments
for _me_ to count instead of just giving me a one or two digit
answer?



joemooreaterolsdotcom

More about : computers count

March 9, 2005 3:44:01 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Wed, 09 Mar 2005 12:44:00 GMT, Joe Moore <munged@bad.example.com>
wrote:

>Computers are supposed to be pretty good at counting things.
>
>So why does my Samsung SIR-T451 digital STB, when I ask it
>to tell me the signal strength of a particular channel, present
>me with a picture of a group of closely spaced line segments
>for _me_ to count instead of just giving me a one or two digit
>answer?
>
>

Because there is no such thing as 100 or whatever you wish the top to
be.
Thumper
>
>joemooreaterolsdotcom

To reply drop XYZ in address
Anonymous
March 9, 2005 8:35:40 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

Thumper <jaylsmithXYZ@comcast.net> wrote:

>On Wed, 09 Mar 2005 12:44:00 GMT, Joe Moore <munged@bad.example.com>
>wrote:
>
>>Computers are supposed to be pretty good at counting things.
>>
>>So why does my Samsung SIR-T451 digital STB, when I ask it
>>to tell me the signal strength of a particular channel, present
>>me with a picture of a group of closely spaced line segments
>>for _me_ to count instead of just giving me a one or two digit
>>answer?
>>
>>
>
>Because there is no such thing as 100 or whatever you wish the top to
>be.

That kind of makes sense, but aren't they implying an upper limit
with the picture, since only so many lines will fit in the picture?

But regardless, wouldn't it make sense to mark a scale with say
0, MINIMUM, OPTIMUM, MAXIMUM?

joemooreaterolsdotcom
Related resources
March 9, 2005 8:35:41 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

"Joe Moore" <munged@bad.example.com> wrote in message
news:0hbu21dd6n7idacuiar6ciurhrbpahnmur@4ax.com...
> Thumper <jaylsmithXYZ@comcast.net> wrote:
>
>>On Wed, 09 Mar 2005 12:44:00 GMT, Joe Moore <munged@bad.example.com>
>>wrote:
>>
>>>Computers are supposed to be pretty good at counting things.
>>>
>>>So why does my Samsung SIR-T451 digital STB, when I ask it
>>>to tell me the signal strength of a particular channel, present
>>>me with a picture of a group of closely spaced line segments
>>>for _me_ to count instead of just giving me a one or two digit
>>>answer?
>>>
>>>
>>
>>Because there is no such thing as 100 or whatever you wish the top to
>>be.
>
> That kind of makes sense, but aren't they implying an upper limit
> with the picture, since only so many lines will fit in the picture?
>
> But regardless, wouldn't it make sense to mark a scale with say
> 0, MINIMUM, OPTIMUM, MAXIMUM?
>

Couldn't you tell just by looking at the bars as to whether you're low,
medium or high??
why do you need a number? A quick look and you could determine if you're
above half or 3/4

much like a fuel gauge in a car... you don't nee any digits to tell you how
much is left in the tank, a quick glance and you know if it's time to refill
or not...
Anonymous
March 9, 2005 11:20:00 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

"Mike" <mike@poolsters.on.ca> wrote:

>
>"Joe Moore" <munged@bad.example.com> wrote in message
>news:0hbu21dd6n7idacuiar6ciurhrbpahnmur@4ax.com...
>> Thumper <jaylsmithXYZ@comcast.net> wrote:
>>
>>>On Wed, 09 Mar 2005 12:44:00 GMT, Joe Moore <munged@bad.example.com>
>>>wrote:
>>>
>>>>Computers are supposed to be pretty good at counting things.
>>>>
>>>>So why does my Samsung SIR-T451 digital STB, when I ask it
>>>>to tell me the signal strength of a particular channel, present
>>>>me with a picture of a group of closely spaced line segments
>>>>for _me_ to count instead of just giving me a one or two digit
>>>>answer?
>>>>
>>>>
>>>
>>>Because there is no such thing as 100 or whatever you wish the top to
>>>be.
>>
>> That kind of makes sense, but aren't they implying an upper limit
>> with the picture, since only so many lines will fit in the picture?
>>
>> But regardless, wouldn't it make sense to mark a scale with say
>> 0, MINIMUM, OPTIMUM, MAXIMUM?
>>
>
>Couldn't you tell just by looking at the bars as to whether you're low,
>medium or high??

Unfortunately, it's not that easy. I've never seen anything lower than
7 bars (except 0). 7 bars gives me dropouts, 8 bars gives me a good
picture, and I haven't seen anything higher than 9.

It just seems to me that distinguishing between 7,8, and 9 bars could
be a lot easier.

>why do you need a number? A quick look and you could determine if you're
>above half or 3/4

Just above half is equivalent to zero. 3/4 means maybe yes and maybe
no. Not a lot of useful information there.

>much like a fuel gauge in a car... you don't nee any digits to tell you how
>much is left in the tank, a quick glance and you know if it's time to refill
>or not...

I think I'm dealing with the equivalent of a fuel guage which shows
5/8 when the tank is empty and shows 3/4 when the tank is full.

It's no big deal, just another frustration dealing with the worlds
balkiest user interface in the SIR-T451.



joemooreaterolsdotcom
March 9, 2005 11:20:01 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
wrote:

>I think I'm dealing with the equivalent of a fuel guage which shows
>5/8 when the tank is empty and shows 3/4 when the tank is full.

No you aren't. What is full as far as signal strength is concerned?

The gas tank has a finite number of gallons. No one knows what 100%
signal strength would be,
Thumper
To reply drop XYZ in address
Anonymous
March 9, 2005 11:20:01 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

Joe Moore (munged@bad.example.com) wrote in alt.tv.tech.hdtv:
> >Couldn't you tell just by looking at the bars as to whether you're low,
> >medium or high??
>
> Unfortunately, it's not that easy. I've never seen anything lower than
> 7 bars (except 0). 7 bars gives me dropouts, 8 bars gives me a good
> picture, and I haven't seen anything higher than 9.

That's because the bars usually represent corrected error count mixed
in with signal strength, where "10" would be zero errors and a *lot* of
excess signal (headroom). The problem is that as soon as the maximum
correctable error threshold per unit time is reached (this number is based
on the amount of FEC in the signal), some errors won't be able to be
corrected and you will get visible (or audible) errors.

The problem is that this happens a *long* time before your signal level
is so low that you can't lock onto a digital carrier.

So, no matter what the method of presenting the data, it's actually pretty
meaningless, because you don't know what weight is being given to error
count and to signal strength. I've seen many STBs where the display is
0 bars for no carrier, and 5-7 bars for carrier but the maximum number of
correctable errors per unit time. If they pick 6 bars for that setting,
then 1-6 bars are all about the same when it comes to your viewing
experience.

--
Jeff Rife |
| http://www.nabs.net/Cartoons/RhymesWithOrange/Obedience...
Anonymous
March 10, 2005 2:04:58 AM

Archived from groups: alt.tv.tech.hdtv (More info?)

Thumper <jaylsmithXYZ@comcast.net> wrote:

>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>wrote:
>
>>I think I'm dealing with the equivalent of a fuel guage which shows
>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>
>No you aren't. What is full as far as signal strength is concerned?

Full (optimum) is high enough to assure the receiver operates as
designed with no errors caused by low signal level.



joemooreaterolsdotcom
March 10, 2005 2:04:59 AM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
wrote:

>Thumper <jaylsmithXYZ@comcast.net> wrote:
>
>>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>>wrote:
>>
>>>I think I'm dealing with the equivalent of a fuel guage which shows
>>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>>
>>No you aren't. What is full as far as signal strength is concerned?
>
>Full (optimum) is high enough to assure the receiver operates as
>designed with no errors caused by low signal level.

measured in what units?
thumper
>
>
>joemooreaterolsdotcom

To reply drop XYZ in address
Anonymous
March 10, 2005 10:23:20 AM

Archived from groups: alt.tv.tech.hdtv (More info?)

Thumper wrote:
> On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
> wrote:
>
>
>>Thumper <jaylsmithXYZ@comcast.net> wrote:
>>
>>
>>>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>>>wrote:
>>>
>>>
>>>>I think I'm dealing with the equivalent of a fuel guage which shows
>>>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>>>
>>>No you aren't. What is full as far as signal strength is concerned?
>>
>>Full (optimum) is high enough to assure the receiver operates as
>>designed with no errors caused by low signal level.
>
>
> measured in what units?
> thumper
>

Well, according to what he wrote, the unit of measure is errors. The
desired number being zero.

--
Matthew

I'm a contractor. If you want an opinion, I'll sell you one.
Which one do you want?
March 10, 2005 6:45:19 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Thu, 10 Mar 2005 07:23:20 -0500, "Matthew L. Martin"
<nothere@notnow.never> wrote:

>Thumper wrote:
>> On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
>> wrote:
>>
>>
>>>Thumper <jaylsmithXYZ@comcast.net> wrote:
>>>
>>>
>>>>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>>>>wrote:
>>>>
>>>>
>>>>>I think I'm dealing with the equivalent of a fuel guage which shows
>>>>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>>>>
>>>>No you aren't. What is full as far as signal strength is concerned?
>>>
>>>Full (optimum) is high enough to assure the receiver operates as
>>>designed with no errors caused by low signal level.
>>


>>
>> measured in what units?
>> thumper
>>
>
>Well, according to what he wrote, the unit of measure is errors. The
>desired number being zero.


That's not what he wrote.
Thumper
To reply drop XYZ in address
Anonymous
March 10, 2005 6:51:39 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

Thumper wrote:
> On Thu, 10 Mar 2005 07:23:20 -0500, "Matthew L. Martin"
> <nothere@notnow.never> wrote:
>
>
>>Thumper wrote:
>>
>>>On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
>>>wrote:
>>>
>>>
>>>
>>>>Thumper <jaylsmithXYZ@comcast.net> wrote:
>>>>
>>>>
>>>>
>>>>>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>>>>>wrote:
>>>>>
>>>>>
>>>>>
>>>>>>I think I'm dealing with the equivalent of a fuel guage which shows
>>>>>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>>>>>
>>>>>No you aren't. What is full as far as signal strength is concerned?
>>>>
>>>>Full (optimum) is high enough to assure the receiver operates as
>>>>designed with no errors caused by low signal level.
>>>
>
>
>>>measured in what units?
>>>thumper
>>>
>>
>>Well, according to what he wrote, the unit of measure is errors. The
>>desired number being zero.
>
>
>
> That's not what he wrote.

Actually it is a restatement of what he wrote, not a quotation. I didn't
quote him nor did I claim to quote him.

--
Matthew

I'm a contractor. If you want an opinion, I'll sell you one.
Which one do you want?
March 10, 2005 9:52:23 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Thu, 10 Mar 2005 15:51:39 -0500, "Matthew L. Martin"
<nothere@notnow.never> wrote:

>Thumper wrote:
>> On Thu, 10 Mar 2005 07:23:20 -0500, "Matthew L. Martin"
>> <nothere@notnow.never> wrote:
>>
>>
>>>Thumper wrote:
>>>
>>>>On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
>>>>wrote:
>>>>
>>>>
>>>>
>>>>>Thumper <jaylsmithXYZ@comcast.net> wrote:
>>>>>
>>>>>
>>>>>
>>>>>>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>>>>>>wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>I think I'm dealing with the equivalent of a fuel guage which shows
>>>>>>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>>>>>>
>>>>>>No you aren't. What is full as far as signal strength is concerned?
>>>>>
>>>>>Full (optimum) is high enough to assure the receiver operates as
>>>>>designed with no errors caused by low signal level.
>>>>
>>
>>
>>>>measured in what units?
>>>>thumper
>>>>
>>>
>>>Well, according to what he wrote, the unit of measure is errors. The
>>>desired number being zero.
>>
>>
>>
>> That's not what he wrote.
>
>Actually it is a restatement of what he wrote, not a quotation. I didn't
>quote him nor did I claim to quote him.
nonsense
Thumper
To reply drop XYZ in address
Anonymous
March 10, 2005 10:21:23 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

Thumper wrote:
> On Thu, 10 Mar 2005 15:51:39 -0500, "Matthew L. Martin"
> <nothere@notnow.never> wrote:
>
>
>>Thumper wrote:
>>
>>>On Thu, 10 Mar 2005 07:23:20 -0500, "Matthew L. Martin"
>>><nothere@notnow.never> wrote:
>>>
>>>
>>>
>>>>Thumper wrote:
>>>>
>>>>
>>>>>On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
>>>>>wrote:
>>>>>>
>>>>>>Full (optimum) is high enough to assure the receiver operates as
>>>>>>designed with no errors caused by low signal level.
>>>>>
>>>
>>>>>measured in what units?
>>>>>thumper
>>>>>
>>>>
>>>>Well, according to what he wrote, the unit of measure is errors. The
>>>>desired number being zero.
>>>
>>>
>>>
>>>That's not what he wrote.
>>
>>Actually it is a restatement of what he wrote, not a quotation. I didn't
>>quote him nor did I claim to quote him.
>
> nonsense

Once again you insist on demonstrating you ignorance and pigheadedness
on usenet. You must make your mama proud.

Matthew

--
Thermodynamics and/or Golf for dummies: There is a game
You can't win
You can't break even
You can't get out of the game
Anonymous
March 11, 2005 5:29:16 AM

Archived from groups: alt.tv.tech.hdtv (More info?)

Thumper <jaylsmithXYZ@comcast.net> wrote:

>On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
>wrote:
>
>>Thumper <jaylsmithXYZ@comcast.net> wrote:
>>
>>>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>>>wrote:
>>>
>>>>I think I'm dealing with the equivalent of a fuel guage which shows
>>>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>>>
>>>No you aren't. What is full as far as signal strength is concerned?
>>
>>Full (optimum) is high enough to assure the receiver operates as
>>designed with no errors caused by low signal level.
>
>measured in what units?

Based on what Mr Rife posted about the display being
a varying combination of signal level and errors in
various time periods, it looks like the bars don't
represent specific quantities of physical units.

But regardless of what the bars represent, it is still
easier to read a number than to count closely spaced parallel line
segments with no background and no scale. The number represents
whatever (apparently somewhat arbitrary) units the bars represent.




joemooreaterolsdotcom
Anonymous
March 11, 2005 5:45:01 AM

Archived from groups: alt.tv.tech.hdtv (More info?)

Jeff Rife <wevsr@nabs.net> wrote:

>Joe Moore (munged@bad.example.com) wrote in alt.tv.tech.hdtv:
>> >Couldn't you tell just by looking at the bars as to whether you're low,
>> >medium or high??
>>
>> Unfortunately, it's not that easy. I've never seen anything lower than
>> 7 bars (except 0). 7 bars gives me dropouts, 8 bars gives me a good
>> picture, and I haven't seen anything higher than 9.
>
>That's because the bars usually represent corrected error count mixed
>in with signal strength, where "10" would be zero errors and a *lot* of
>excess signal (headroom). The problem is that as soon as the maximum
>correctable error threshold per unit time is reached (this number is based
>on the amount of FEC in the signal), some errors won't be able to be
>corrected and you will get visible (or audible) errors.
>
>The problem is that this happens a *long* time before your signal level
>is so low that you can't lock onto a digital carrier.
>
>So, no matter what the method of presenting the data, it's actually pretty
>meaningless, because you don't know what weight is being given to error
>count and to signal strength. I've seen many STBs where the display is
>0 bars for no carrier, and 5-7 bars for carrier but the maximum number of
>correctable errors per unit time. If they pick 6 bars for that setting,
>then 1-6 bars are all about the same when it comes to your viewing
>experience.

Thanks for the informative explanation. I've given up my quest
for meaning in bars. I still think that since the number of bars
has some significance if only in comparison to smaller and larger
numbers of bars, that presenting an actual number instead of (or in
addition to) a group of bars would make the display more useful for
repetitious reading and adjustments. Especially when there is some
time or distance between the adjustment and the display.




joemooreaterolsdotcom
Anonymous
March 11, 2005 5:45:02 AM

Archived from groups: alt.tv.tech.hdtv (More info?)

Joe Moore (munged@bad.example.com) wrote in alt.tv.tech.hdtv:
> I still think that since the number of bars
> has some significance if only in comparison to smaller and larger
> numbers of bars, that presenting an actual number instead of (or in
> addition to) a group of bars would make the display more useful for
> repetitious reading and adjustments.

Yeah, you are right on this. But, since the number really won't help
diagnose any problems, that's probably why they didn't bother.

The MyHD software that runs the MIT PCI HDTV cards has a program that
displays actual error rate count (corrected and uncorrected) as well as
carrier-to-noise ratio. The software with DVICO's Fusion cards displays
raw signal strength. With all this, I can pretty much see what the exact
problem is, and it's about the only way to know what really is going on
without a spectrum analyzer.

--
Jeff Rife |
| http://www.nabs.net/Cartoons/RhymesWithOrange/Obedience...
March 11, 2005 4:35:14 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Thu, 10 Mar 2005 19:21:23 -0500, "Matthew L. Martin"
<nothere@notnow.never> wrote:

>Thumper wrote:
>> On Thu, 10 Mar 2005 15:51:39 -0500, "Matthew L. Martin"
>> <nothere@notnow.never> wrote:
>>
>>
>>>Thumper wrote:
>>>
>>>>On Thu, 10 Mar 2005 07:23:20 -0500, "Matthew L. Martin"
>>>><nothere@notnow.never> wrote:
>>>>
>>>>
>>>>
>>>>>Thumper wrote:
>>>>>
>>>>>
>>>>>>On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
>>>>>>wrote:
>>>>>>>
>>>>>>>Full (optimum) is high enough to assure the receiver operates as
>>>>>>>designed with no errors caused by low signal level.
>>>>>>
>>>>
>>>>>>measured in what units?
>>>>>>thumper
>>>>>>
>>>>>
>>>>>Well, according to what he wrote, the unit of measure is errors. The
>>>>>desired number being zero.
>>>>
>>>>
>>>>
>>>>That's not what he wrote.
>>>
>>>Actually it is a restatement of what he wrote, not a quotation. I didn't
>>>quote him nor did I claim to quote him.
>>
>> nonsense
>
>Once again you insist on demonstrating you ignorance and pigheadedness
>on usenet. You must make your mama proud.
>
>Matthew

Get lost dopey.
Thumper
To reply drop XYZ in address
March 11, 2005 4:39:10 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Fri, 11 Mar 2005 02:29:16 GMT, Joe Moore <munged@bad.example.com>
wrote:

>Thumper <jaylsmithXYZ@comcast.net> wrote:
>
>>On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
>>wrote:
>>
>>>Thumper <jaylsmithXYZ@comcast.net> wrote:
>>>
>>>>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>>>>wrote:
>>>>
>>>>>I think I'm dealing with the equivalent of a fuel guage which shows
>>>>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>>>>
>>>>No you aren't. What is full as far as signal strength is concerned?
>>>
>>>Full (optimum) is high enough to assure the receiver operates as
>>>designed with no errors caused by low signal level.
>>
>>measured in what units?
>
>Based on what Mr Rife posted about the display being
>a varying combination of signal level and errors in
>various time periods, it looks like the bars don't
>represent specific quantities of physical units.
>
>But regardless of what the bars represent, it is still
>easier to read a number than to count closely spaced parallel line
>segments with no background and no scale. The number represents
>whatever (apparently somewhat arbitrary) units the bars represent.
>
>
True but I think you understand what I'm saying. If a number of 50%
comes up it has to be mean 50% of something. The bars indicate
relative strength based on the arbitrary scale it has selected. A
number scale might be better if it actually had a reference.
Thumper
>
>
>joemooreaterolsdotcom

To reply drop XYZ in address
Anonymous
March 12, 2005 4:05:54 AM

Archived from groups: alt.tv.tech.hdtv (More info?)

Thumper <jaylsmithXYZ@comcast.net> wrote:

>On Fri, 11 Mar 2005 02:29:16 GMT, Joe Moore <munged@bad.example.com>
>wrote:
>
>>Thumper <jaylsmithXYZ@comcast.net> wrote:
>>
>>>On Wed, 09 Mar 2005 23:04:58 GMT, Joe Moore <munged@bad.example.com>
>>>wrote:
>>>
>>>>Thumper <jaylsmithXYZ@comcast.net> wrote:
>>>>
>>>>>On Wed, 09 Mar 2005 20:20:00 GMT, Joe Moore <munged@bad.example.com>
>>>>>wrote:
>>>>>
>>>>>>I think I'm dealing with the equivalent of a fuel guage which shows
>>>>>>5/8 when the tank is empty and shows 3/4 when the tank is full.
>>>>>
>>>>>No you aren't. What is full as far as signal strength is concerned?
>>>>
>>>>Full (optimum) is high enough to assure the receiver operates as
>>>>designed with no errors caused by low signal level.
>>>
>>>measured in what units?
>>
>>Based on what Mr Rife posted about the display being
>>a varying combination of signal level and errors in
>>various time periods, it looks like the bars don't
>>represent specific quantities of physical units.
>>
>>But regardless of what the bars represent, it is still
>>easier to read a number than to count closely spaced parallel line
>>segments with no background and no scale. The number represents
>>whatever (apparently somewhat arbitrary) units the bars represent.
>>
>>
>True but I think you understand what I'm saying.

I understand, I just disagree that your argument outweighs
the legibility argument.

> If a number of 50%
>comes up it has to be mean 50% of something. The bars indicate
>relative strength based on the arbitrary scale it has selected.

50% could represent half the distance along that same arbitrary scale.

> A
>number scale might be better if it actually had a reference.

The problem with that view is that no matter how arbitrary the
scale is, a group of bars is still a representation of a number
to the human mind. Otherwise, folks wouldn't refer to their signal
strength using expressions like "8 bars".

I concede that "8 bars" has no absolute meaning except in comparison
to "7 bars" and "9 bars" on the same device.

Will you concede that the number 8 or even the expression "8 bars" is
more _legible_ than a group of closely spaced parallel lines segments
which must be counted to extract the exact same (admittedly relative
and not absolute) meaning?

Any unwarranted conclusions about absolute meaning are not likely to
affect the task for which this presentation is designed, namely making
adjustments to optimize the performance on a particular channel.





joemooreaterolsdotcom
March 12, 2005 12:04:06 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Sat, 12 Mar 2005 01:05:54 GMT, Joe Moore <munged@bad.example.com>
wrote:

>Will you concede that the number 8 or even the expression "8 bars" is
>more _legible_ than a group of closely spaced parallel lines segments
>which must be counted to extract the exact same (admittedly relative
>and not absolute) meaning?


Sure but the world is full of graphs. For relative strength I don't
see why numbers are needed. Numbers imply a measurement of some
specific unit.
Thumper
To reply drop XYZ in address
Anonymous
March 12, 2005 7:10:00 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

If I remember correctly(it's been about five years since I had DirecTV)
signal strength was shown as both a number AND a bar. Obviously a better
way to go!

Bill

"Jeff Rife" <wevsr@nabs.net> wrote in message
news:MPG.1c9ad59bfe5f004f989c1e@news.nabs.net...
Joe Moore (munged@bad.example.com) wrote in alt.tv.tech.hdtv:
> I still think that since the number of bars
> has some significance if only in comparison to smaller and larger
> numbers of bars, that presenting an actual number instead of (or in
> addition to) a group of bars would make the display more useful for
> repetitious reading and adjustments.

Yeah, you are right on this. But, since the number really won't help
diagnose any problems, that's probably why they didn't bother.

The MyHD software that runs the MIT PCI HDTV cards has a program that
displays actual error rate count (corrected and uncorrected) as well as
carrier-to-noise ratio. The software with DVICO's Fusion cards displays
raw signal strength. With all this, I can pretty much see what the
exact
problem is, and it's about the only way to know what really is going on
without a spectrum analyzer.

--
Jeff Rife |
|
http://www.nabs.net/Cartoons/RhymesWithOrange/Obedience...
Anonymous
March 12, 2005 8:27:12 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

Thumper <jaylsmithXYZ@comcast.net> wrote:

>On Sat, 12 Mar 2005 01:05:54 GMT, Joe Moore <munged@bad.example.com>
>wrote:
>
>>Will you concede that the number 8 or even the expression "8 bars" is
>>more _legible_ than a group of closely spaced parallel lines segments
>>which must be counted to extract the exact same (admittedly relative
>>and not absolute) meaning?
>
>
>Sure but the world is full of graphs.

But the world is not full of graphs which show the equivalent of
a single digit in comparison to nothing.

> For relative strength I don't
>see why numbers are needed.

If the display actually did show the relative strength
of the current measurement to the previous one, I would
agree with this. But what is actually displayed is only
current strength. A single digit's worth of information.
The display relies on the observer's memory of a graph
seen an unspecified time ago to determine the relative
change necessary to validate adjustments. That's why
people count the bars.

> Numbers imply a measurement of some
>specific unit.

So do graphs but so what? How does the implication that
a specific unit is being shown affect the task which the
display is supposed to aid? I contend that it doesn't hurt
and that a number is easier to compare to previous measurements.


joemooreaterolsdotcom
March 12, 2005 8:27:13 PM

Archived from groups: alt.tv.tech.hdtv (More info?)

On Sat, 12 Mar 2005 17:27:12 GMT, Joe Moore <munged@bad.example.com>
wrote:

>Thumper <jaylsmithXYZ@comcast.net> wrote:
>
>>On Sat, 12 Mar 2005 01:05:54 GMT, Joe Moore <munged@bad.example.com>
>>wrote:
>>
>>>Will you concede that the number 8 or even the expression "8 bars" is
>>>more _legible_ than a group of closely spaced parallel lines segments
>>>which must be counted to extract the exact same (admittedly relative
>>>and not absolute) meaning?
>>
>>
>>Sure but the world is full of graphs.
>
>But the world is not full of graphs which show the equivalent of
>a single digit in comparison to nothing.
>

Whatever
Goodbye.
Thumper
>> For relative strength I don't
>>see why numbers are needed.
>
>If the display actually did show the relative strength
>of the current measurement to the previous one, I would
>agree with this. But what is actually displayed is only
>current strength. A single digit's worth of information.
>The display relies on the observer's memory of a graph
>seen an unspecified time ago to determine the relative
>change necessary to validate adjustments. That's why
>people count the bars.
>
>> Numbers imply a measurement of some
>>specific unit.
>
>So do graphs but so what? How does the implication that
>a specific unit is being shown affect the task which the
>display is supposed to aid? I contend that it doesn't hurt
>and that a number is easier to compare to previous measurements.
>
>
>joemooreaterolsdotcom

To reply drop XYZ in address
!