Hi Everyone, What are the different ways to avoid CPU overload to get the best recording. I am using a 2009 MBP 2.53 Ghz Inter core 2 duo, 8 GB ram running 64 bit Logic Pro on 32 bit snow leopard kernel. I use an external 7200 rpm USB 2.0 hard drive for accessing external software plug ins. What i know so far 1) Increase buffer size 2) Freeze other tracks to reduce CPU usage. anything else?

Right off the bat, ditch the USB drive and get a FW800 drive. Then get yet another FW800 for any sample-based plugins. Also... projects should be saved/played back from a project drive and never the internal drive. In short, you want one hard drive "per task" for optimum performance: - system drive: never used to save/play back Logic projects. That's the task of a... - project drive - separate HD's for sample libraries - never use USB drives for anything other than backup. Use only FW or eSATA (not applicable with an MPB because I don't believe there are any 3rd party adapters you can get to provide an eSATA port)

Pros and cons of other methods to prevent overloads: 1) as you mentioned, higher buffer size. But for playing in parts, that can become a pain. The higher the latency (which is what happens when you increase the buffer size), the harder it is to play in time. The only solution there is to not use software monitoring and use an alternative form of monitoring (hardware mixer, or, an interface that provides "direct monitoring" capability). 2) Freezing Tracks. Can be counterproductive if the number of tracks you're freezing exceed a certain threshold, and that threshold can only be determined on a case-by-case, project-by-project basis. Freeze files are 32 bit, so they take up more room on a HD than a normal 24-bit audio file. Thus, if you're playing back a lot of frozen tracks, each one will consume 25% more bandwidth to read back than a normal audio file. ALTERNATIVE: bounce your tracks to normal 24 bit files. Not as elegant as freezing, but there ya go. ;-) 3) Getting a more powerful computer. I know, I know, who has the cash?!!! But it's a viable option if the density/requirements of your productions are large.

Ski, thanks for the comments you said for optimum performance use separate hard drives for sample libraries and another hard drive for storing and recording logic songs. Let's say I am getting only one external hard drive(FW800) for logic recordings and for sample libraries. Is that still okay :-) what about the sample libraries that were installed with logic which are currently in my system drive. Should i move them as well? For a technical understanding how is the performance improving when everything in different hard drive vs all in the system drive?

[b]Let's say I am getting only one external hard drive(FW800) for logic recordings and for sample libraries. Is that still okay :-)[/b] Nope! :-) Not a good idea. With that setup, if you're playing back audio tracks (especially frozen tracks), the more samples you start using by adding sample-based instruments, the worse your performance will get. [b]what about the sample libraries that were installed with logic which are currently in my system drive. Should i move them as well?[/b] Theoretically, yes. But for all practical purposes, no. Even though I generally stress having separate hard drives for all sample libraries, you'll be just fine leaving the stock Logic stuff where it is. [b]For a technical understanding how is the performance improving when everything in different hard drive vs all in the system drive?[/b] The short answer: with one drive trying to do all the work you tax the drive's capacity to deliver all the data you might need. Drives don't have an unlimited capacity to read and write data in real time. Depending on the density of your productions, you can easily exceed a drive's ability to deliver all the data you need when it's all located on a single drive. So when you have multiple drives you decrease the possibility of overloads because the data is coming from several different sources.

Thanks for the simple explanation. How does increasing buffer size improve performance. What is actually happening. Also I noticed that when the CPU overloads the amount of RAM used is still quite low. I am guessing now this is mainly because the hard drive must have reached its limit. But i would love for you to explain where RAM play a role and where the processor speed plays a role with respect all things done with logic.

Hi Anand, I have a less powerful system than you have. Here is one tip--and I have no idea why this works but it does--include in your project an empty audio track, no files no plugins. Select this track and then play back your project. What I have found is that this simple trick tends to spread the processing more evenly over the cores. Again I have no idea why this works.

You're welcome. Increasing buffer size... The processor(s) in your computer, while very fast, still need a certain amount of time to process everything you have going on in your projects. Now, the following is an over-simplification, but I think it'll get the point across: Did you ever watch the I Love Lucy episode where she's working in the chocolate factory? Go to this link and start watching from about 4:04... http://www.youtube.com/watch?v=GLp7Y4TxXSA So now, the less chocolaty version! The I/O buffer values are in samples. So with that in mind... One stereo audio track with an EQ on it is going to require a certain amount of time to process raw audio (as it lives on your hard drive) and play it back processed with the EQ. For argument's sake, let's say this takes 10 sample's worth of time to accomplish. If your buffer size was set to 32 then you'd probably be fine. Now you have, say, 10 stereo tracks, each with an EQ. So now your computer needs at least 100 samples worth of time to read all the data from your hard drive, process it, and spit it all back out. So you probably need a buffer size of 128, at least, to prevent the system from reporting that it can't process all the data in time ("overload"). Having the buffer set too high can lead to problems too, but the general rule of thumb is that you want to set your buffer size as small as possible without sacrificing the performance of your system. Again, this is an oversimplification, but that's the theory.

Thanks Ski My MBP has one firewire 800 port and 2 USB port. I use a USB hub on one port and connect my external hard disc which contains the samples through the other one directly. If i get two hard discs with firewire800, i need to get a firewire hub. Now won't that essentially half my speed. would it not be better to use one 480 mbps USB port and another 800 mbps firewire port.

Let me clear up a couple of things: 1) FW800 is faster than USB2.0. See this [url]http://docs.info.apple.com/article.html?artnum=75471[/url] Apple document for reference. 2) You won't have to get a FW hub. You can daisy-chain your two FW800 drives and get exactly the same performance. 3) To the best of my understanding, aside from FW800 being faster, USB drives cannot handle the same degree of sustained (constant) data transfer to/from the drive as FW800 drives. 4) When you connect two FW800 in a daisy chain or otherwise (e.g., using a hub) the transfer speed of data along the bus does not get cut in half. Post back if you need more clarification.