Prev: Since I'm already on the run I can start stirring police again...
Next: monitor purchase
From: Rod Speed on 6 Jul 2010 06:28
John Tserkezis wrote

>>>> Ehhhmm... a total drop of 1.2 - 1.4V on *each* battery
>>>> voltage of 1.5V seems quite a show stopper to me....

>>> Doesnt have to be anything like that much of a drop.

> Pick any Schotkky diode. .1something to .4something volts drop
> depending on flavour. Though, IMO even that would be too much.

> You need to appreciate that this type of cost for the
> feature you're getting, is way more than gross luxury.

Wrong when its designed in to the device right from the start.

> So technically it's certainly very possible, but
> the economics will be below ordinary at best,

Wrong when its designed in to the device right from the start.

> and broke at worst.

Wrong when its designed in to the device right from the start.

> When you're trying to drill that last few cents out of
> a product that's destined for serious mass production
> at a minimal cost, every fraction of a cent counts.

Have fun explaining the led thats included anyway etc.

> It's just not worth it.

Thanks for that completely superfluous proof of why no one is
actually stupid enough to employ you to design anything that matters.


From: Andrew Smallshaw on 6 Jul 2010 07:36
On 2010-07-06, Rod Speed <rod.speed.aaa(a)gmail.com> wrote:
> John Tserkezis wrote
>
>> You need to appreciate that this type of cost for the
>> feature you're getting, is way more than gross luxury.
>
> Wrong when its designed in to the device right from the start.

Yes, even when it's designed right in from the start. Do you want
to change the habit of a lifetime and start _justifying_ your
pronouncements instead of simply endlessly repeating them as if
that alone is enough to make them true?

>> So technically it's certainly very possible, but
>> the economics will be below ordinary at best,
>
> Wrong when its designed in to the device right from the start.

Yes, even when it's designed into the device right at the start.
Not that it really matters, this kind of circuitry is basically
invisible to the rest of the system aside from any voltage drop.
You can put it in at the start or before laying out the final
production board - it doesn't make that much difference

>> and broke at worst.
>
> Wrong when its designed in to the device right from the start.

Yes, even when it's designed into the device right at the start.

>> When you're trying to drill that last few cents out of
>> a product that's destined for serious mass production
>> at a minimal cost, every fraction of a cent counts.
>
> Have fun explaining the led thats included anyway etc.

Because a cost/benefit analysis (however informal) shows that LED
is worth including. It's a standard design trade off, cost vs.
functionality. For some devices, those indicator LEDs are the
_only_ sign of life that is not dependent on connected equipment.

>> It's just not worth it.
>
> Thanks for that completely superfluous proof of why no one is
> actually stupid enough to employ you to design anything that matters.

It was more of a proof than simply spouting "not when it's designed
in right at the start" in parrot-like fashion. The way I'd do this
would use four transistors and four resistors, plus a bit of board
space, extra soldering, possibly extra drilling, more faults etc.
I don't see it costing much less than about 8p even with a reasonable
production run. For some sectors that is unacceptable even on
equipment going for three figures. If the device is supposed to
sell for a fiver it is unacceptable anywhere.

--
Andrew Smallshaw
andrews(a)sdf.lonestar.org
From: Rod Speed on 6 Jul 2010 14:36
Andrew Smallshaw wrote
> Rod Speed <rod.speed.aaa(a)gmail.com> wrote
>> John Tserkezis wrote

>>> You need to appreciate that this type of cost for the
>>> feature you're getting, is way more than gross luxury.

> Wrong when its designed in to the device right from the start.

> Yes, even when it's designed right in from the start.

Wrong, as always. It doesnt cost enough to matter
when the extra is included in the special purpose ic.

> Do you want to change the habit of a lifetime and start _justifying_
> your pronouncements instead of simply endlessly repeating them
> as if that alone is enough to make them true?

You're so stupid that it isnt worth the trouble.

>>> So technically it's certainly very possible, but
>>> the economics will be below ordinary at best,

>> Wrong when its designed in to the device right from the start.

> Yes, even when it's designed into the device right at the start.

Wrong, as always. It doesnt cost enough to matter
when the extra is included in the special purpose ic.

> Not that it really matters,

It happens to be what is being discussed.

> this kind of circuitry is basically invisible to the rest of the
> system aside from any voltage drop. You can put it in at
> the start or before laying out the final production board
> - it doesn't make that much difference

It makes a considerable difference when its all in a special purpose ic.

>>> and broke at worst.

>> Wrong when its designed in to the device right from the start.

> Yes, even when it's designed into the device right at the start.

Wrong, as always. It doesnt cost enough to matter
when the extra is included in the special purpose ic.

>>> When you're trying to drill that last few cents out of
>>> a product that's destined for serious mass production
>>> at a minimal cost, every fraction of a cent counts.

>> Have fun explaining the led thats included anyway etc.

> Because a cost/benefit analysis (however informal) shows that LED is worth including.

Just as true of allowing the batterys to go in any way the user likes.

> It's a standard design trade off, cost vs. functionality.

So his original claim is just plain wrong, as I said.

> For some devices, those indicator LEDs are the _only_
> sign of life that is not dependent on connected equipment.

And they are included anyway even when they arent.

So much for his stupid claim.

>>> It's just not worth it.

>> Thanks for that completely superfluous proof of why no one is
>> actually stupid enough to employ you to design anything that matters.

> It was more of a proof than simply spouting "not when
> it's designed in right at the start" in parrot-like fashion.

Corse you never ever do anything like that yourself, eh ?

> The way I'd do this would use four transistors and four resistors, plus a bit
> of board space, extra soldering, possibly extra drilling, more faults etc.

Anyone with even half a clue would include whats needed in the special purpose ic.

> I don't see it costing much less than about 8p even with a reasonable production run.

And it wouldnt cost anything like that when its included in the special purpose ic.

> For some sectors that is unacceptable even on equipment going for three figures.
> If the device is supposed to sell for a fiver it is unacceptable anywhere.

Have fun explaining how the absolute vast bulk of those have a led or lcd.


From: Andrew Smallshaw on 6 Jul 2010 16:19
On 2010-07-06, Rod Speed <rod.speed.aaa(a)gmail.com> wrote:
>
> Wrong, as always. It doesnt cost enough to matter
> when the extra is included in the special purpose ic.

So, the 99.9+% of designs (including, for example, most computer
motherboards) that use no custom ASICs are a complete irrelevence,
are they?

> Wrong, as always. It doesnt cost enough to matter
> when the extra is included in the special purpose ic.

Adding _power_ transistors to your typical ASIC will certianly not
be free. I suspect you would be looking long and hard for a foundry
to even entertain the idea. It is competely impossible with the
sea of gates ASICs for a start.

> Andrew Smallshaw wrote
>> Not that it really matters,
>
> It happens to be what is being discussed.

It was your assertion that when it is designed in is somehow pivotal
to how much it costs. If you had continued to read the very
sentence you truncated you would have seen that _that_ makes no
real difference.

>> Because a cost/benefit analysis (however informal) shows that
>> LED is worth including.
>
> Just as true of allowing the batterys to go in any way the user
> likes.

The user is accustomed to ensuring battery polarities are correct.
How many devices out there have this kind of any-way-will-do
circuitry? If there was a massive demand for it it would have been
addressed long ago.

>> I don't see it costing much less than about 8p even with a
>> reasonable production run.
>
> And it wouldnt cost anything like that when its included in the
> special purpose ic.

No, of course it wouldn't cost anything like that. Instead it
would probably be at least a capital cost of �100,000 for the ASIC
and another �1 per unit to accommodate those on chip power transistors.

>> For some sectors that is unacceptable even on equipment going
>> for three figures. If the device is supposed to sell for a fiver
>> it is unacceptable anywhere.
>
> Have fun explaining how the absolute vast bulk of those have a
> led or lcd.

I already have. You chose to invent a new economic reality instead
of reading it. Now I remember why you were in my killfile.

--
Andrew Smallshaw
andrews(a)sdf.lonestar.org
From: Rod Speed on 6 Jul 2010 17:00
Andrew Smallshaw wrote
> Rod Speed <rod.speed.aaa(a)gmail.com> wrote
>> Andrew Smallshaw wrote
>>> Rod Speed <rod.speed.aaa(a)gmail.com> wrote
>>>> John Tserkezis wrote

>>>>> You need to appreciate that this type of cost for the
>>>>> feature you're getting, is way more than gross luxury.

>>> Wrong when its designed in to the device right from the start.

>>> Yes, even when it's designed right in from the start.

>> Wrong, as always. It doesnt cost enough to matter
>> when the extra is included in the special purpose ic.

> So, the 99.9+% of designs (including, for example, most
> computer motherboards) that use no custom ASICs

Those use ASICs designed for motherboards, stupid.

> are a complete irrelevence, are they?

They certainly are to the BATTERY POWERED devices being discussed.

>>>>> So technically it's certainly very possible, but
>>>>> the economics will be below ordinary at best,

>>>> Wrong when its designed in to the device right from the start.

>>> Yes, even when it's designed into the device right at the start.

>> Wrong, as always. It doesnt cost enough to matter
>> when the extra is included in the special purpose ic.

> Adding _power_ transistors to your typical ASIC will certianly not be free.

No one said a word about free except you.

> I suspect you would be looking long and hard for a foundry to even entertain
> the idea. It is competely impossible with the sea of gates ASICs for a start.

Even someone as stupid as you should have noticed that the BATTERY
POWERED devices actually being discussed dont actually use those much.

>>> Not that it really matters,

>> It happens to be what is being discussed.

> It was your assertion that when it is designed in is somehow pivotal to how much it costs.

Everyone can see for themselves that I said nothing like that.

> If you had continued to read the very sentence you truncated

I did that, and replied to that bit as well.

> you would have seen that _that_ makes no real difference.

Wrong, as always. It doesnt cost enough to matter
when the extra is included in the special purpose ic.

>>> this kind of circuitry is basically invisible to the rest of the
>>> system aside from any voltage drop. You can put it in at
>>> the start or before laying out the final production board
>>> - it doesn't make that much difference

>> It makes a considerable difference when its all in a special purpose ic.

>>> When you're trying to drill that last few cents out of
>>> a product that's destined for serious mass production
>>> at a minimal cost, every fraction of a cent counts.

>> Have fun explaining the led thats included anyway etc.

>>>>> When you're trying to drill that last few cents out of
>>>>> a product that's destined for serious mass production
>>>>> at a minimal cost, every fraction of a cent counts.

>>>> Have fun explaining the led thats included anyway etc.

>>> Because a cost/benefit analysis (however informal) shows that LED is worth including.

>> Just as true of allowing the batterys to go in any way the user likes.

> The user is accustomed to ensuring battery polarities are correct.

And that patent was about allowing the user to ignore that
and allows for little kids not needing to be taught that etc.

> How many devices out there have this kind of any-way-will-do circuitry?

Irrelevant to the obvious advantage with that approach.

> If there was a massive demand for it it would have been addressed long ago.

The same stupid claim could have been made about all
sorts of things that have only recently become common.

>>> I don't see it costing much less than about 8p even with a reasonable production run.

>> And it wouldnt cost anything like that when its included in the special purpose ic.

> No, of course it wouldn't cost anything like that. Instead it would
> probably be at least a capital cost of �100,000 for the ASIC and

Not when the device needs that already, fool.

> another �1 per unit to accommodate those on chip power transistors.

Not when the device has some already, fool.

>>> For some sectors that is unacceptable even on equipment
>>> going for three figures. If the device is supposed to sell
>>> for a fiver it is unacceptable anywhere.

>> Have fun explaining how the absolute vast bulk of those have a led or lcd.

> I already have.

Like hell you have.

> You chose to invent a new economic reality instead of reading it.

Everyone can see you are lying, as always.

> Now I remember why you were in my killfile.

Put me back, then we wont have to see any more of your pathetic excuse for mindless bullshit.


First  |  Prev  | 
Pages: 1 2 3
Prev: Since I'm already on the run I can start stirring police again...
Next: monitor purchase