logarithmic scale

    Dec 07 2010 | 3:18 pm
    hello : )
    I have used the 'scale' object before which is great, but I wonder if there is a logarithmic scale object under another name i have missed that has preset response curves or a curve that can be plotted on a graph by the user allowing for a scale result that is not so linear.
    The other alternative I see is to make lots of individual splits and scale objects to create a logarithmic result, but I was hoping for a more efficient and tidy solution.perhaps someone has done something like this before?
    thanks for any help. : )

    • Dec 07 2010 | 3:30 pm
      Hello teqy,
      the [scale] object, can use an exponential base to have a more "curved" effect other than a linear output! may you could use it for what you want!
    • Dec 07 2010 | 4:16 pm
      There are a few Max objects that will 'do' log maths, the simplest being [scale], [pow] and [expr]:
    • Dec 07 2010 | 4:29 pm
      [expr] also has log10 and log (natural log).
      If you want to draw your mapping by hand you can use [table].
    • Dec 07 2010 | 5:24 pm
      Hi teqy,
      Here's an abstraction that you may find helpful.
      best, Zachary
    • Dec 14 2010 | 3:03 pm
      Zachary - this object is great. Thanks for sharing..
    • Jan 17 2011 | 5:35 pm
      about time that scale object got patched. tanx alot!
    • Jan 17 2011 | 7:36 pm
      Thanks srs and dtr. Glad you like it. I have a similar version of zmap, and other abstractions here: http://www.zacharyseldess.com/z.abstractions.html
      best, Zachary
    • Jun 14 2013 | 3:44 am
      Thank you Zachary for the abstractions!
    • Jun 14 2013 | 8:23 am
      FYI, there is a thread that has been started a while ago with the intention of summmarizing simple mapping methods that maxers know about: https://cycling74.com/forums/re-mapping-a-range-of-numbers-add-your-tricks-here
      Cheers, Ádám
    • Feb 09 2014 | 3:57 pm
      Thank you Zachary, just what I needed! A newbie question: are these objects written in C?
    • Apr 02 2017 | 5:00 am
      Thanks Zachary- Of course your z.scale is exactly what I was looking for.
      Any ideas why the standard scale object is so hard to use (the 5th argument log part)? I see it has been updated, and there is now a classic verse modern version of the log implementation (toggle-able via attribute)- but couldnt get the standard log curve I was after out of it no matter what I did with it. The curve I was after was the one I get with your z.scale using 2. as the log value (as per your help patch).
      Just adding this post as it seems likely they fixed up the standard scale object, I just still cant work it out...