Convert 1024 to 1KB like "%.2p" SI conversion

[skol45uk]

Hi

Is there a function or feature that will convert an integer or double value to an SI equivalent decimal string, but using 1024 as the divisor? 

I.e

2^10 = 1024 (K)

2^20 = 1048576 (M)

2^30 =  1073741824 (G)

Etc

In other words -

0 < N < 1024 = "N"

1024 < N < 1048576 = "<N/1024> K"

1048576 < N < 1073741824 = "<N/1048576> M"

ETC

I'd be surprised if such a function does not exist. Or am I being naive?

Thanks

[AeroSoul]

I don't know if it exists, but it's really not hard to do by yourself:

[JÞB]

Use an array of "",k,M,G,T and look up quotient of Log2(x),10 as the index.

Format "%f %sB" where f is Remainder log2(x), 10 and %s is the string from the Lookup table

[paul_cardinale]

---

@AeroSoul wrote:

I don't know if it exists, but it's really not hard to do by yourself:

---

That won't do it.

This kind of works, but the result is funky if the input is < 1024.

[AeroSoul]

This works for practically anything, but i don't really like how it handles the decimal points.

JPB's idea, but i couldn't get his to work with the Remainder Log2(x), 10, so i did x/(1024^(log2(x)/10))

 

[JÞB]

---

@AeroSoul wrote:

This works for practically anything, but i don't really like how it handles the decimal points.

JPB's idea, but i couldn't get his to work with the Remainder Log2(x), 10, so i did x/(1024^(log2(x)/10))

 

---

Well, that would be because I failed to describe the conversion correctly.  CERTAINLY this was a problem with trying to type from my phone and not a failure to properly think through the math 🙂

I don't understand why x is not a Unsigned Integer in AeroSoul's picture of a snippet Last I checked Bytes can't be negative or fractional.

[JÞB]

Minor mod to add coverage for Peta and Exa (Zetta and Yotta are ignored since a U64 0xFFFFFFFFFFFFFFFF is a mere16EB-1Byte.)  We need P and E to prevent indexing past the array size and making large numbers look small

[AeroSoul]

---

@JÞB wrote:

I don't understand why x is not a Unsigned Integer in AeroSoul's picture of a snippet Last I checked Bytes can't be negative or fractional.

---

Because right click--> create control on the log2(x) makes it a double 😄