Skip navigation
This discussion is archived

Macbook C2D and Shake 4.1??

1406 Views 15 Replies Latest reply: Dec 22, 2006 4:19 PM by Inconceivable RSS
1 2 Previous Next
Reynmaker Calculating status...
Currently Being Moderated
Nov 17, 2006 9:18 PM
I know this may sound silly but is there anyone out there that's currently running shake 4.1 on the new MacBooks? I'm still in the learning phase so I'm planning on bringing a portable to work and do the tutorials in my free time.

Mac Pro 2.6 | 1TB HD | 2GB RAM | 2x23" / G5 2x2.0 | 3GB RAM   Mac OS X (10.4.7)   FCP Studio, After Effects 7, Shake 4.1, Decklink PCIe
  • aneks Calculating status...
    Currently Being Moderated
    Nov 17, 2006 10:42 PM (in response to Reynmaker)
    I would not recommend using shake on any system that does not have a decent OpenGL card as it is so dependant on openGL for everything it does. If you are havn't bought already a macbook pro is a much option
    Mac Pro, Mac OS X (10.4.8), dual xeon 2.66 4gb ram 7300gt
  • Patrick Sheffield Level 7 Level 7 (29,175 points)
    Currently Being Moderated
    Nov 19, 2006 11:59 AM (in response to aneks)
    That's only for the GUI, isn't it. It doesn't use OPENGL acceleration for rendering, does it?

    Patrick

    Dual 2 GHz G5, Mac OS X (10.4.7), 2.5 GB RAM, nVidia 6800 Ultra, Decklink Extreme
  • Eugene Debs Level 1 Level 1 (95 points)
    Currently Being Moderated
    Nov 20, 2006 12:39 AM (in response to Reynmaker)
    I know this may sound silly but is there anyone out
    there that's currently running shake 4.1 on the new
    MacBooks? I'm still in the learning phase so I'm
    planning on bringing a portable to work and do the
    tutorials in my free time.

    Mac Pro 2.6 | 1TB HD | 2GB RAM |
    2x23" / G5 2x2.0 | 3GB RAM   Mac OS X (10.4.7)
      FCP Studio, After Effects 7, Shake 4.1,
    Decklink PCIe


    Runs great. My 15" MBP usually runs faster than my PPC dual-2 GHz G5. The 17" is even better. One little tiny refresh problem with on-screen controls when you modify them with a slider, but nothing to lose sleep over.

    -ed
  • Patrick Sheffield Level 7 Level 7 (29,175 points)
    Currently Being Moderated
    Nov 20, 2006 2:01 PM (in response to Reynmaker)
    I honestly don't think Shake uses OpenGL for rendering... just the GUI.

    Patrick

    Dual 2 GHz G5, Mac OS X (10.4.7), 2.5 GB RAM, nVidia 6800 Ultra, Decklink Extreme
  • Johnny Sharman Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 26, 2006 2:51 PM (in response to Reynmaker)
    I recently bought a 2.33 macbook pro core 2 duo and am interested in shake on it - anyone got any firm views in addition to those already here?

    Johnny
    MBP C2D 2.33, Mac OS X (10.4.8)
  • aneks Level 1 Level 1 (20 points)
    Currently Being Moderated
    Nov 26, 2006 9:37 PM (in response to Johnny Sharman)
    Yes,you are right Patrick. OpenGl is an interaction issue only. rendering is just number crunching and will not draw any gpu power.

    Shake should perform very tolerably on a c2d macbook pro, especially if you have 2gb or more ram !
    Mac Pro, Mac OS X (10.4.8), dual xeon 2.66 4gb ram 7300gt
  • Eisen Feuer Calculating status...
    Currently Being Moderated
    Nov 28, 2006 10:35 AM (in response to Reynmaker)
    I run Shake on my 17" MBP, and it works fine (That is, until you start dabbling in 100 node scripts that take 10 seconds to render a single frame ). I don't see why it would be any different on the Core 2 Duo machines.
    MacBook Pro 17" 2.16GHz+2GB RAM, and a PowerMac G5 Single 1.8GHz+1.5GB RAM, Mac OS X (10.4.7)
  • Johnny Sharman Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 30, 2006 11:29 AM (in response to Eisen Feuer)
    I run Shake on my 17" MBP, and it works fine (That
    is, until you start dabbling in 100 node scripts that
    take 10 seconds to render a single frame ). I
    don't see why it would be any different on the Core 2
    Duo machines.


    Thanks for this - do you mind if i ask you to elaborate a bit - i will be new to and training on shake - i'd really appreciate it if you would tell me more about your experience with laptop. is it basically running just ok(ish) on mbp - do the limitations appear all too quickly - or is 100 nodes quite a complicated procedure. REALLY GRATEFUL for your help. Cheers Johnny
    MBP C2D 2.33, Mac OS X (10.4.8)
  • Patrick Sheffield Level 7 Level 7 (29,175 points)
    Currently Being Moderated
    Nov 30, 2006 1:32 PM (in response to Johnny Sharman)
    Personally, I think if you've got a hundred nodes, you may be doing something wrong - or at least you might try altering your work flow...



    This is one of the more complicated scripts I've done and it is less than 50, I believe...

    Patrick

    Dual 2 GHz G5, Mac OS X (10.4.7), 2.5 GB RAM, nVidia 6800 Ultra, Decklink Extreme
  • Eisen Feuer Level 1 Level 1 (20 points)
    Currently Being Moderated
    Nov 30, 2006 6:47 PM (in response to Johnny Sharman)
    I don't have time to write the story of the experiences with my MBP, the best way to get that across would be conversation style- AIM: EisenFeuer7 MSN/Gtalk: eisenfeuer@gmail.com

    And Patrick, this is a medium-complexity script right here (of the eye I linked earlier), and this is one of the few scripts involving a simple almost static object.

    does this work?

    I'm working on a vast almost hollywood-like project right now that I can't link you to (typical secret production stuff) but anyway, the shots involve filters that I've constructed in Shake and those filters, sometimes 25 nodes a piece, converge in a singe script to make a truly massive piece of rendering. I just learned what proxies are (by mistake, actually) so that should help some, but I'm still wondering if there are any corners you can cut- maybe specific cases of quality limiters on the big hitter nodes?

    MacBook Pro 17" 2.16GHz+2GB RAM, and a PowerMac G5 Single 1.8GHz+1.5GB RAM   Mac OS X (10.4.7)  
  • aneks Level 1 Level 1 (20 points)
    Currently Being Moderated
    Dec 7, 2006 4:14 PM (in response to Eisen Feuer)
    Frankly. These is not a very big scripts at all, I am not trying to start a contest here, but many Shake users do much more complicated shots on a daily basis . I can see from the picture you have posted that there a really large number of poor workflow decisions and un-neccesary nodes in this tree. You could dramatically optimise this scene by taking advantage of things like concatenation of transforms and colour corections as well thinmking through your comp a little better.

    I really suggest you at least flick through the mannual and have a look at the section on good work practises. I am not trying to be rude or critical but if you where one of my students or the compositors on my team I would make you rebuild that tree with about a third the number of nodes.
    Mac Pro, Mac OS X (10.4.8), dual xeon 2.66 4gb ram 7300gt
  • Eisen Feuer Level 1 Level 1 (20 points)
    Currently Being Moderated
    Dec 8, 2006 2:14 AM (in response to aneks)
    well, I WISH I was a student of someone because right now besides the few questions I've asked here, I'm self-taught.

    And yes, I've seen big scripts from Weta Digital for Lord of the Rings, but for a normal guy like me working at home rendering with a laptop instead of a render farm, these can get pretty big.

    What page is the section you're talking about- I searched the pdf manual in with every conceivable variant of "good work practices" and I'm finding nothing.
    MacBook Pro 17" 2.16GHz+2GB RAM, and a PowerMac G5 Single 1.8GHz+1.5GB RAM, Mac OS X (10.4.7)
  • aneks Level 1 Level 1 (20 points)
    Currently Being Moderated
    Dec 22, 2006 3:58 PM (in response to Eisen Feuer)
    Please take these comments as constructive criticism as it appears you are still getting to know shake. In a big production environment senior artists will generally gsteer you clear of mistakes.

    three big things I am seeing in the script. you are using a non-concatenating colour node right at the top of the script. This is bad. then you are stacking a couple of masked fade nodes with non masked nodes. Basically I never use the mask input an any node. It's bad too, break concatenation and generally messy. You seem to be doing a lot of glowing type effects by using brightness and blur and over. Mayeb try using screen or add, thereby saving you a numbr of brightness nodes. Then you stack brightness and fade nodes in a non-concatenating fashion which is esssentially a redundant calculation.

    Also you doubly break concatentaion on one branch where you use an adjustHSV then a blur then a bunch of concatenating colour corrects. On the left hand side of the image you seem to be masking a file-in node which is also not good practice.

    You seem to be using film grain to generate and effect upstream but a noise or and funciton here will probably be a better choice as you are not actually grain matching. You have 3 wapers and an rblur in this tree which will slow anything down. Try pre-rendering some of the warpers. Avoid film grain and rBlur except when absolutely necessary.

    Take advantage of colour and transform concatentation. Avoid using use the under node as it is redundant and only there for legacy code. Don't use the mask input on stacked colour corrections.
    Mac Pro, Mac OS X (10.4.8), dual xeon 2.66 4gb ram 7300gt
1 2 Previous Next

Actions

More Like This

  • Retrieving data ...

Bookmarked By (0)

Legend

  • This solved my question - 10 points
  • This helped me - 5 points
This site contains user submitted content, comments and opinions and is for informational purposes only. Apple disclaims any and all liability for the acts, omissions and conduct of any third parties in connection with or related to your use of the site. All postings and use of the content on this site are subject to the Apple Support Communities Terms of Use.