Michael Stokowski:

Michael Stokowski:

• Always accept “3k4” notation as 3.4k (no longer optional)

Does this mean I will have to accept the DOT madness as used by American engineers ?

I gather you disapprove the 3.4k (with a dot) notation. Here in France we academically write 3,4k (with a comma), although everybody understand 3k4, which is colloquial and not correct at all.

My understanding is this new version understands 3k4 as 3.4k.
I don't think it does with 3,4k.

I wrote:

This discussion is very confused, but for more than a century, it has been normal practice for the decimal point (dot) to be used in Britain, the Far East and the USA, and the decimal comma to be used in Europe and some other countries. Even ISO and IEC have not been able to resolve this.

It would be good for LTspice to allow 3,4k as well as 3.4k and 3k4. The latter notation is surely quite easy to understand, The decimal multiplier symbol takes the place of the decimal point or comma, so for example 1µ0, 2p7, 4G7.

I wrote:

>Le 07/11/2024 à 15:46, Goran Finnberg via groups.io a écrit :

>Michael Stokowski:

• Always accept “3k4” notation as 3.4k (no longer optional)

I wrote: Does this mean I will have to accept the DOT madness as used by American engineers ?

Jerry Lee Marcel wrote:

I gather you disapprove the 3.4k (with a dot) notation. 

Here in France we academically write 3,4k (with a comma), 

As I do in Sweden.

We from Europe NEVER EVER use the DOT instead we use the comma.

>although everybody understand 3k4, which is colloquial and not correct at all.

And why is that so ?

What does k stand for ?

1 kHz ?

1 kohm ?

2k4 ?

What is the difference between the above 3 ?

Why exclude 2k4 as it is equally valid as, 1 kohm or 1 kHz ?

>My understanding is this new version understands 3k4 as 3.4k.


Then I would like to have that confirmed by Michael Stokowski.

>I don't think it does with 3,4k.

Then 3.4k is wrong from the perspective of European engineers.

Why do I have to accept the DOT madness ?

--------------

Best regards,

Goran Finnberg
The Mastering Room AB
Goteborg
Sweden

E-mail: mastering@...

Learn from the mistakes of others, you can never live long enough to
make them all yourself.    -   John Luther

(\__/)
(='.'=)
(")_(") Pyret, Ranglet, Aron, VovVov, Nero, Smurfen & Pussin:RIP

-- 
OOO - Own Opinions Only
Best Wishes
John Woodgate
Keep trying

Well, most of the anglo-saxons use dot as decimal separator. Is that mad? they could say that we are mad to use comma as a separator.
The choice is historical. Both are understandable.
The fact that ISO has opted for comma doesn't make those that use dot villains.
If it did, all people that do not speak teh same language would be villains.

although everybody understand 3k4, which is colloquial and not correct at all.

And why is that so ?

What does k stand for ?

1 kHz ?

1 kohm ?

2k4 ?

What is the difference between the above 3 ?

Being pedantic, only "2k4" is correct in SPICE context. The other two have a space between the numeric and the multiplier, so the multiplier and everything that follows is either ignored, or causes a syntax error in LTspice, depending on where it is used.

BTW, the comma isn't universally used in Europe. It is not recognised in the UK, Ireland, Switzerland or Liechtenstein. And in good old Luxembourg, both the comma and the dot are accepted as alternatives.

Personally, I try to never use any decimal point or comma in an electrical context. There are decadic multipliers that make that unnecessary. I learned that lesson decades ago when all drawings went through a photocopier before being distributed. Somewhere at the bottom of all those machines are piles of lost decimal points. It's a sad state that different sectors cannot  agree on the format of decadic multipliers. Financial contexts often use "mn" to signify million - £107mn, for example.

Le Système international d’unités (SI units) details decadic multipliers from 10²⁴ (yotta, Y) through to 10^-24 (yocto, y) that should be capable of succinctly fulfilling every conceivable number in popular use. LTspice supports a sub-set: all from T (Tera) down to f (femto) in steps of 1000x (deca, hecto, deci and centi are not recognised). Peta ("P", 10¹⁵) is interpreted as "p" (pico), as SPICE is case-agnostic. Atto (10^-18) and below are also not recognised and are ignored.

--
Regards,
Tony

Did you try it? Curiously, on a machine set to use the decimal point, LTspice interprets '3,4k' as 4000 ohms, not 3 ohms, as one might expect..

On Thu, 7 Nov 2024 at 17:04, John Woodgate via groups.io <jmw=woodjohn.uk@groups.io> wrote:

>It would be good for LTspice to allow 3,4k as well as 3.4k 

This should already happen using regionalisation of the OS.

-- 
OOO - Own Opinions Only
Best Wishes
John Woodgate
Keep trying

>It would be good for LTspice to allow 3,4k as well as 3.4k

I don't think so.

From SPICE3 docu:
"Fields on a line are separated by one or more blanks, a comma, an equal (’=’) sign, or a left or right
parenthesis."

For compatibility reasons with SPICE3, and probably other SPICE versions, the comma must therefore remain a separator.

Bernhard

Le Système international d’unités (SI units) details decadic multipliers from 10²⁴ (yotta, Y) through to 10^-24 (yocto, y) that should be capable of succinctly fulfilling every conceivable number in popular use. LTspice supports a sub-set: all from T (Tera) down to f (femto) in steps of 1000x (deca, hecto, deci and centi are not recognised). Peta ("P", 10¹⁵) is interpreted as "p" (pico), as SPICE is case-agnostic. Atto (10^-18) and below are also not recognised and are ignored.

The problem is that the decadic multipliers are in steps of 1000.
For example a 90.9k resistor would be written as 909e2. That is not very intuitive when reading a schematic. It requires an additional effort compared to the system most of us have grown accustomed with. I believe ther's the same risk of misreading as between 90900 and 909000.

LTspice differs from SPICE3 in many ways, so one more wouldn't matter much. But my finding that 3,4k is interpreted as 4000 needs investigating, I think.

Actually, SPICE3 can be criticised for wasting valuable symbols: it should have specified just one separator (or a minimal hierarchy for nesting), rather than an undifferentiated, and surely incomplete, list.

>It would be good for LTspice to allow 3,4k as well as 3.4k

I don't think so.

From SPICE3 docu:
"Fields on a line are separated by one or more blanks, a comma, an equal (’=’) sign, or a left or right
parenthesis."

For compatibility reasons with SPICE3, and probably other SPICE versions, the comma must therefore remain a separator.

Bernhard

-- 
OOO - Own Opinions Only
Best Wishes
John Woodgate
Keep trying

The problem is that the decadic multipliers are in steps of 1000.
For example a 90.9k resistor would be written as 909e2. That is not very intuitive when reading a schematic. It requires an additional effort compared to the system most of us have grown accustomed with. I believe ther's the same risk of misreading as between 90900 and 909000.

No, it would be written as 90k9. Although 909e2 is legal and would be understood by SPICE. I didn't mean for the term "decadic multipliers" to be synonym for scientific format.

It is also common for manufacturers of passive component parts to use creative formats in order to avoid the use of decimal points in part numbers, e.g. C0402C333xxxxx is Kemet's naming convention for a 33n capacitor in a 0603 case size (imperial units).

--