1 Reply Latest reply: Mar 19, 2010 5:37 AM by ssprengel RSS

    Lightroom and processing platform question

    Monsieur Septimus Fry Community Member

      I have posted a version of this question on PS, but it is LR where it really matters.

       

      Some eons ago, I was a PS user and found my then-current machine had features which PS was not using. Cannot remember the fine detail, we are talking PS5 or 6...

       

      So, now I am contemplating buying a new i7 based machine with umpteen GB of memory and super graphics board. All these raise questions:can LRuse the speed features of the i7, can they use the umpteen GB, switch into the graphics stuff.

       

      I am fed up with the P4 3.2GHz, 3GB, GeForce 9800GT platform I have - it stumbles and makes me wait (.always Loading... taking seconds). will I end up finding things not much better on a platform potentially 50 times faster? Or is 50 times faster just a dream?

       

      If someone replies that their xxx with yyy goes like lightning, they will not be actually answering the question. I want to know if LR (and PS) can use all the bells and whistles available to them as processes - are they 64-bit so able to use the nnGB... etc etc

       

      Thanks...

        • 1. Re: Lightroom and processing platform question
          ssprengel Community Member

          LR and PS can use multiple cores, perhaps not all 6 if you get a $1000 CPU--I base that on a benchmark that had LR running the same on both a quad and hex-core although no details were available on what the LR portion of the benchmark was doing.  LR and PS both have 64-bit editions. 

           

          CS4-PS takes advantage of the GPU for certain things, but LR seems to only be sped up by the OS using the GPU more, not LR using the GPU, specifically, itself. 

           

          CS5 and LR3 are just around the corner and we don't know everything GPU-wise they will take advantage of, although there are reports that CS5-Premier Pro will only use CUDA (nVidia) not OpenCL (ATI) because OpenCL is buggy, still.

           

          50-times-faster would be a benchmark that can use all the cores to their fullest, whereas LR does some things in parallel but not nearly everything, so the speedup will vary depending on what you do.

           

          I am about ready to build myself a i7 960 machine and expect LR and PS to be very fast on it.  I am not sure if I'll try using an SSD or not for the OS.  If I go with what I'm leaning towards at the moment, a Gigabyte GA X58A-UD7, it would likely have USB3 on a couple of the USB ports with a 6Gbps speed that a few peripherals can use.  It also has SATA3, which is half that speed, but faster than SATA2.  I don't know as I will be overclocking it, much, which depends on the cooling setup, so maybe the MB is overkill and I'll go for something cheaper.

          http://www.gigabyte.us/Products/Motherboard/Products_Overview.aspx?ProductID=3344