Swap level being ignored

On 11/5/2010 1:25 PM, Brett Holden wrote:

I have a library components (dual FET, Si4834BDY) that has two electrically

halves, yet the pin names are the same and number of pins are same for the

two halves. One FET has a schottky diode in parallel in from S-D, and one

FET does not.

 

I choose to use swap level 1 for each half, even though they are internally

different. The external pinouts are mirrors of each other.

 

What is the problem in allowing me to swap the two halves, so long as there

are similar pin assignments?

 

My sugggestion is, unless there is some commanding reason why you cannot do

so, then HONOR a request for swap-level = 1, and swap two electrically

different device halves - so long as the pin-assignments are the same, even

if those halves use two different symbols!

 

>

you could make a single gate, each pair of pins swappable and then

pinswap all 3 in turn, each pin having a different swap level...but then

you are stuck with grouping the transistors together on the schematic.

I agree with your suggestion, it's none of EAGLE's business if your

gates don't match, at most a warning "Swapped gates are not equivalent"

"Gary Gofstein" <nospam@use.forum.net> wrote in message

news:ib215s$m6v$1@cheetah.cadsoft.de...

On 11/5/2010 1:25 PM, Brett Holden wrote:

>> snip

you could make a single gate, each pair of pins swappable and then pinswap

all 3 in turn, each pin having a different swap level...but then you are

stuck with grouping the transistors together on the schematic.

I agree with your suggestion, it's none of EAGLE's business if your gates

don't match, at most a warning "Swapped gates are not equivalent"

Gary, thanks for your suggestion, but "-ouch-".

I got to thinking about this after I aired my beef... .

It was a lost cause since before dinosaurs and beer.

Minor annoyance. I fix.

I was annoyed because I spent about 20 minutes trying to repair my library.

It took me that long before I realized the problem was systemic. Maybe I got

off easy.

"Brett Holden" <bretth2o@bellsouth.net> wrote in message

news:ibilb2$lf8$1@cheetah.cadsoft.de...

 

"Gary Gofstein" <nospam@use.forum.net> wrote in message

news:ib215s$m6v$1@cheetah.cadsoft.de...

>> On 11/5/2010 1:25 PM, Brett Holden wrote:

>>> snip

>> you could make a single gate, each pair of pins swappable and then

>> pinswap all 3 in turn, each pin having a different swap level...but then

>> you are stuck with grouping the transistors together on the schematic.

>> I agree with your suggestion, it's none of EAGLE's business if your gates

>> don't match, at most a warning "Swapped gates are not equivalent"

 

Gary, thanks for your suggestion, but "-ouch-".

I got to thinking about this after I aired my beef... .

It was a lost cause since before dinosaurs and beer.

Minor annoyance. I fix.

 

I was annoyed because I spent about 20 minutes trying to repair my library.

It took me that long before I realized the problem was systemic. Maybe I got

off easy.

"Brett Holden" <bretth2o@bellsouth.net> wrote in message

news:ibilb2$lf8$1@cheetah.cadsoft.de...

 

"Gary Gofstein" <nospam@use.forum.net> wrote in message

news:ib215s$m6v$1@cheetah.cadsoft.de...

>> On 11/5/2010 1:25 PM, Brett Holden wrote:

>>> snip

>> you could make a single gate, each pair of pins swappable and then

>> pinswap all 3 in turn, each pin having a different swap level...but then

>> you are stuck with grouping the transistors together on the schematic.

>> I agree with your suggestion, it's none of EAGLE's business if your gates

>> don't match, at most a warning "Swapped gates are not equivalent"

 

Gary, thanks for your suggestion, but "-ouch-".

I got to thinking about this after I aired my beef... .

It was a lost cause since before dinosaurs and beer.

Minor annoyance. I fix.

 

"Brett Holden" wrote:

>I have a library components (dual FET, Si4834BDY) that has two electrically

>halves, yet the pin names are the same and number of pins are same for the

>two halves. One FET has a schottky diode in parallel in from S-D, and one

>FET does not.

>I choose to use swap level 1 for each half, even though they are internally

>different. The external pinouts are mirrors of each other.

>What is the problem in allowing me to swap the two halves, so long as there

>are similar pin assignments?

>My sugggestion is, unless there is some commanding reason why you cannot do

>so, then HONOR a request for swap-level = 1, and swap two electrically

>different device halves - so long as the pin-assignments are the same, even

>if those halves use two different symbols!

since many years, my suggestion is to allow any swap on demand, IOW

instead of the error message, there should be a dialog offering "swap

anyway", maybe with some more information (see older postings, maybe

in the German groups).

And swaps should be possible from the board! Pinswaps are trivial,

gateswap should highlight the pins belonging to the gate.

Oliver

On 11/11/2010 11:30 PM, Oliver Betz wrote:

"Brett Holden" wrote:

 

>> I have a library components (dual FET, Si4834BDY) that has two electrically

>> halves, yet the pin names are the same and number of pins are same for the

>> two halves. One FET has a schottky diode in parallel in from S-D, and one

>> FET does not.

>>

>> I choose to use swap level 1 for each half, even though they are internally

>> different. The external pinouts are mirrors of each other.

>>

>> What is the problem in allowing me to swap the two halves, so long as there

>> are similar pin assignments?

>>

>> My sugggestion is, unless there is some commanding reason why you cannot do

>> so, then HONOR a request for swap-level = 1, and swap two electrically

>> different device halves - so long as the pin-assignments are the same, even

>> if those halves use two different symbols!

 

since many years, my suggestion is to allow any swap on demand, IOW

instead of the error message, there should be a dialog offering "swap

anyway", maybe with some more information (see older postings, maybe

in the German groups).

 

And swaps should be possible from the board! Pinswaps are trivial,

gateswap should highlight the pins belonging to the gate.

 

Oliver

yeah, and pinswaps would be nicer if they would swap the pins on the

schematic instead of the nets, especially if initiated in board view.