Maker Pro
Maker Pro

Question About IC Chips

J

John Fields

Jan 1, 1970
0
Obviously there are cases where the extra cent won't make it worthwhile.

But consider this: what if YOU are not the programmer working on the
device? Things change, people change. It's very likely that firmware will
be changed by someone else, someone who may not have the experience,
someone who test benches new firmware, notes everything is fine, releases
the firmware to the field, and all of a sudden hundreds of dead devices
are being returned. Why? The specific conditions that set the port to zero
(when tied to one, or vice versa) were never seen during this
inexperienced programmers testing.

---
If I've designed a piece of equipment for a client, and it works like
it's supposed to, then I've done my job and everyone is happy. If,
then, someone comes along behing me and starts trying to make changes
without knowing what they're doing, it certainly doesn't reflect
badly on me, it reflects badly on whoever was supervising the idiot
programmer.
---
Consider the costs then??

---
Why should I care? I didn't have anything to do with the failures.
---
Or consider this: MCUs aren't infallible, do ugly things to the power
rails or expose them to ESD and it's very possible for the port direction
control bits to flip. Even strong RF can do it.

---
So what? If those are eventualities which are to be expected in the
field, then the proper time to address them is during the design, and
if it's prudent to add pullups or pulldowns it should be done as part
of the process, but certainly not blindly, and _cedrtainly_ not
because some idiot may come along and screw with your code.
---
 
J

John Woodgate

Jan 1, 1970
0
I read in alt.binaries.schematics.electronic that John Fields
The damage to the company's repuation would have come from the idiot
programmer's work not having been supervised closely enough, not from
any hardware design fault on my part.

But you've moved on, whereas he has a clear field to explain to the CEO
that 'The designer didn't take standard precautions against unforeseen,
but reasonably foreseeable, events. This was a disaster waiting to
happen.'

This thread is spread across six newsgroups. Is that really necessary?
 
J

John Fields

Jan 1, 1970
0
I read in alt.binaries.schematics.electronic that John Fields


But you've moved on, whereas he has a clear field to explain to the CEO
that 'The designer didn't take standard precautions against unforeseen,
but reasonably foreseeable, events. This was a disaster waiting to
happen.'
 
S

Stewart Pinkerton

Jan 1, 1970
0
I read in alt.binaries.schematics.electronic that John Fields


But you've moved on, whereas he has a clear field to explain to the CEO
that 'The designer didn't take standard precautions against unforeseen,
but reasonably foreseeable, events. This was a disaster waiting to
happen.'

This thread is spread across six newsgroups. Is that really necessary?

Sure it is, because Fields is desperate to cover his idiot arse!
*Good* designers think about what might go wrong, not just what the
circuit will do if everything is perfect. It's analogous to software
error-trapping.

As you rightly say, 'standard precautions against reasonably
foreseeable events'.

Stewart Pinkerton - professional hardware design engineer

No, I wouldn't really use a pretentious sig like that, because I
actually am one.......................... :)
 
T

Terry Given

Jan 1, 1970
0
Stewart said:
Sure it is, because Fields is desperate to cover his idiot arse!
*Good* designers think about what might go wrong, not just what the
circuit will do if everything is perfect. It's analogous to software
error-trapping.

As you rightly say, 'standard precautions against reasonably
foreseeable events'.

Stewart Pinkerton - professional hardware design engineer

No, I wouldn't really use a pretentious sig like that, because I
actually am one.......................... :)

I tend to agree with John Fields on this one, mostly because a lot of my
work has had stringent budget constraints. A suitable compromise would
be to attach notes to the schematic and code (eg SFR values) pointing
out the steps taken and the potential ramifications.

A far more likely scenario is that the software doesnt work properly in
the first place. Hardware that cant tolerate software going apeshit is a
definite cause for concern.

AFA SFR bit-flipping is concerned, the paranoid programmer re-loads SFRs
every interrupt, just in case. This precautionary measure, unlike using
resistors, costs only a tiny amount of development time; unit build cost
is zero.

Cheers
Terry
 
J

John Fields

Jan 1, 1970
0
Sure it is, because Fields is desperate to cover his idiot arse!

---
That makes absolutely _no_ sense. If I was wrong about something, how
would I be covering my ass by broadcasting it to an ever-increasing
audience?
---
*Good* designers think about what might go wrong, not just what the
circuit will do if everything is perfect. It's analogous to software
error-trapping.

As you rightly say, 'standard precautions against reasonably
foreseeable events'.

---
Well, dumbfuck, if you'd bothered to have read all of the thread
before firing off one of your mealy-mouthed epithets you might,
perhaps, have noticed that I didn't unequivocally state that pullups
should never be used, I said they should be used when their inclusion
in the circuit is warranted. Simple. waste not, want not.
---
 
R

repatch

Jan 1, 1970
0
---
If I've designed a piece of equipment for a client, and it works like it's
supposed to, then I've done my job and everyone is happy. If, then,
someone comes along behing me and starts trying to make changes without
knowing what they're doing, it certainly doesn't reflect badly on me, it
reflects badly on whoever was supervising the idiot programmer.
---

No, it reflects badly on the company as a whole, and since you are part of
the company it does end up reflecting badly on you.

My God, if everybody thought this way...
---
So what? If those are eventualities which are to be expected in the
field, then the proper time to address them is during the design, and if
it's prudent to add pullups or pulldowns it should be done as part of
the process, but certainly not blindly, and _cedrtainly_ not because
some idiot may come along and screw with your code. ---

It's cheap insurance, like with most insurance it's your choice whether
you want it or not. Since the cost is practically zero I don't see why you
have so much resistance to it.

You know, I've worked with people who have an attitude like yours, and all
I'll say is that person no longer works for the company I work for.

If all you care about is doing it your way, and that your way seems to
work, so be it. The fact that you don't CARE that your way may result in
problems in the future, even though they are not directly your fault...
well that's just scary.
 
R

repatch

Jan 1, 1970
0
Well, dumbfuck, if you'd bothered to have read all of the thread before

Ahh, now I understand. You are one of those people who troll newsgroups
and start flame wars. And then, when backed into a corner where their
point of view is clearly not viable, resorts to foul language and juvenile
insults...

Sorry to everyone else for contributing to this person's sorry game.

And plonk to you Mr. Fields.
 
J

John Woodgate

Jan 1, 1970
0
I read in alt.binaries.schematics.electronic that repatch
Ahh, now I understand. You are one of those people who troll newsgroups
and start flame wars. And then, when backed into a corner where their
point of view is clearly not viable, resorts to foul language and juvenile
insults...

Sorry to everyone else for contributing to this person's sorry game.

And plonk to you Mr. Fields.

Don't rush to judgement in this case. Someone else involved has a
'past'.
 
B

Bill Bowden

Jan 1, 1970
0
One very good designer I knew (he's now deceased) not
Why not just connect the unused inputs to some output, so
the input is terminated, and you don't need the resistor?

-Bill
 
B

Bill Bowden

Jan 1, 1970
0
One very good designer I knew (he's now deceased) not
Why not just connect the unused inputs to some output, so
the input is terminated, and you don't need the resistor?

-Bill
 
R

Roger Johansson

Jan 1, 1970
0
Bill Bowden said:
Why not just connect the unused inputs to some output, so
the input is terminated, and you don't need the resistor?

Let's say I have an unused logic inverter gate, and somewhere close to it
an output which is running at 10MHz.

If I tie the input of the unused gate to the output of the clock signal
the earlier unused gate will now run at 10MHz, consuming a lot of power
which is wasted.

So it is better to tie the unused input to ground or Vcc, which will lock
it at a steady state, which will waste a minimum of power.
 
R

Ross Herbert

Jan 1, 1970
0
Let's say I have an unused logic inverter gate, and somewhere close to it
an output which is running at 10MHz.

If I tie the input of the unused gate to the output of the clock signal
the earlier unused gate will now run at 10MHz, consuming a lot of power
which is wasted.

So it is better to tie the unused input to ground or Vcc, which will lock
it at a steady state, which will waste a minimum of power.

http://www.sxlist.com/techref/logic/xtrapins.htm
 
R

Ross Herbert

Jan 1, 1970
0
More specifically, what part of that long text are you referring to, and why?

Do you know how to select, copy and paste a line of text?


I sure do. BUT, I didn't post a "line of text". I posted a hyperlink
to a site giving pointers as to the do's and dont's of connecting
unused pins on IC's. If your newsreader doesn't show a simple and
fairly short hyperlink correctly, that's not my fault.
 
J

John Fields

Jan 1, 1970
0
If all you care about is doing it your way, and that your way seems to
work, so be it. The fact that you don't CARE that your way may result in
problems in the future, even though they are not directly your fault...
well that's just scary.

---
I'll stand behind everything I do and take full responsibility for it,
but when someone starts messing with my stuff, then the guarantee
expires. It's very simple, but obviously very difficult for crybabies
like you, who need to have their asses continually wiped for them to
understand.
 
J

John Fields

Jan 1, 1970
0
Ahh, now I understand. You are one of those people who troll newsgroups
and start flame wars. And then, when backed into a corner where their
point of view is clearly not viable, resorts to foul language and juvenile
insults...

Sorry to everyone else for contributing to this person's sorry game.

And plonk to you Mr. Fields.
 
R

Roger Johansson

Jan 1, 1970
0
I sure do. BUT, I didn't post a "line of text". I posted a hyperlink
to a site giving pointers as to the do's and dont's of connecting
unused pins on IC's. If your newsreader doesn't show a simple and
fairly short hyperlink correctly, that's not my fault.

As you quoted a text from me and answered with a link I assumed you
wanted to reply to what I said, maybe tell me I was wrong or comment what
I said, so I spent 5 minutes trying to find out what in that text could
be taken as a reply to me, so I was rather irritated when I couldn't find
any sense in anything.

If you had bothered to write a few words by yourself, like:

"More on this subject: <link>"

you would have saved me a lot of unnecessary work.
 
M

Mark Jones

Jan 1, 1970
0
Roger said:
As you quoted a text from me and answered with a link I assumed you
wanted to reply to what I said, maybe tell me I was wrong or comment what
I said, so I spent 5 minutes trying to find out what in that text could
be taken as a reply to me, so I was rather irritated when I couldn't find
any sense in anything.

If you had bothered to write a few words by yourself, like:

"More on this subject: <link>"

you would have saved me a lot of unnecessary work.

Oh go have a drink for christ's sake.
 
R

Ross Herbert

Jan 1, 1970
0
As you quoted a text from me and answered with a link I assumed you
wanted to reply to what I said, maybe tell me I was wrong or comment what
I said, so I spent 5 minutes trying to find out what in that text could
be taken as a reply to me, so I was rather irritated when I couldn't find
any sense in anything.

If you had bothered to write a few words by yourself, like:

"More on this subject: <link>"

you would have saved me a lot of unnecessary work.

Oh come now, surely you jest.

Your post was specifically related to tying unused inputs to some
logic level or other and I added to your post with a link which starts
with the header "What should I do with pins I am not using?" If this
isn't self explanatory then I feel sorry for you.
 
Top