Mac Pro for Music Production - Upgrade CPU, RAM, HD? Overclock??!

hughythomas

hughythomas

hughythomas

Associate
Joined
11 Oct 2006
Posts
1,404
Location
Reigate
Hi guys!

I have the early 2006 Mac Pro which I use exclusively for audio production work (on Logic Pro 7, Pro Tools 8 and Cubase 5 but mostly Logic as this is my preferred DAW). I was wondering if anyone wise knew if you could upgrade the CPU on these things? Or do you have to buy a completely new mac?

I use this computer purely for DAWs. Here are the specs:

2006 Mac Pro running:
2X Dual-Core Intel Xeons @ 2.66Ghz
4GB FB-DIMM DDR2 RAM (@667Mhz)
Samsung Spinpoint F1 1TB (for Music)
Default 250GB mac drive for system disk (don't know what it is)
X1900XT Radeon
2X MOTU 828MKIIs in series on Firewire
UAD-2 Nevana 32 DSP suite

What makes the most difference to DAW performance at low buffer sizes hardware wise? CPU? RAM? HD? I don't want to have to fork out for an entirely new system and am quite happy to upgrade the CPUs myself if it can be done?

Many thanks!

Tom
 
There is an app for the 2006 and 2008 mac pro but its hit and miss and I think it also alters the clock speed (IE a second is no longer a second).

You can however put quad CPUs in there if you find the right ones.

Google and yee shall find.

3GHz quads I think its your limit.
 
The CPU is the main component to keep buffer sizes low, you would only need to increase the ram if you find that your DAW along with your virtual instruments are running out of memory.

You could also look at getting a faster hard drive for your main system drive, but the CPU is the main factor in a DAW machine if you want to keep buffer sizes and latency low.

I'm not too sure if you can upgrade the CPU your self in a Mac Pro, probably can be done but I got a feeling it could turn expensive and could be quite tricky.
 
Great cheers for the replies guys.

To the last poster, do you use Logic in a professional environment? I run Logic 7 @ 96kHz on all my projects and the lowest buffer size I can get for tracking is 64. For mixing, UAD-2 needs a high buffer size so I set 1024 but even the Logic plugins kill the DAW's playback engine if I am doing a rough mix of something at 64.

What was the name of the overclocking program for the Mac?

I'm in desperate need of squeezing as much performance out of this thing before even considering selling it...

Thanks again!

Tom
 
Well, from a technical standpoint 96kHz produces less aliasing and is a truer representation of the original soundwaves. To my ears, it sounds 'fuller' and has an inherent amount of clarity which simply is not present at anything lower. It is difficult to describe in words.

I appreciate that 88.2kHz would be a better choice due to it being a multiple of 44.1kHz. Bearing in mind we have to downsample to this samplerate anyways at mastering it would seem the obvious choice. However, if the mastering suite employs good converters this isn't such a problem and again, to my ears 96kHz just has this 'sound' about it which I can't seem to obtain if I use anything less. I even once tried 192kHz and although I personally couldn't notice any difference the engineer I was working with could - so it really comes down to personal opinion.

P.S The mac died though @ 192kHz lol even just a stereo recording of an acoustic guitar!!!
 
Have you ever blind tested 44.1, 48, 96 and 192kHz? ;) Marketing departments love the numbers game... ;)
 
Yer we blind tested them all not to sound up myself here but I got them all right!!! I've been working in music for a while, the ability to be pitch perfect and hear finite details like sample rates, mp3 compression algorithms all come with time and doing lots of it!!!
 
I record stuff at 96KHz, you can always downsample after if you want, but my DAC needs modding before I can play back 96KHz stuff so I need to break out the soldering iron.

FWIW, I can't hear sine-waves above 16Khz, I done a few tests with a synth and an scope when I was bored... :o

Keep everything 96Khz if you can, then just bounce down to 44.1KHz at the end for the mastering/pressing person. :)

192 won't break machines, but not much gear can manage 192 in the first place, 96Khz is plenty unless you've got the hearing of a dog! ;)
 
I only ask because personally I find the difference minute and not really worth the extra storage space. There's also Nyquists, which suggests 44.1kHz is already overkill. Compression - yes I can hear that and I'm sure most people can when played back to back in a blind test, but sample rates... You're a one of a few in my experience. :)

Mike - personally I can hear up to about 18k, albeit only just.
 
You got to remember the most important part of recording (in my opinion) is capturing the original source sound as accurately as possible. Assuming of course you are not using Fisher Price microphones and Tomy instruments, 96kHz can make a shed load of difference as it is a truer digital representation of the analogue source.
 
I can't hear above 15-16kHz myself in terms of frequency range which sucks. I'm sure had I not done all the gigging/live performances in the past I would be OK but working in music inevitably destroys your hearing. I myself suffer from tinnitus 24/7 even now that I use proper moulded ear protection :( :(
 
Fair enough - just for further discussion though...

I don't see how 96kHz can make a shed load of difference.

Even the best mics in the world don't warrant the need for 96khz sample rates. A Neumann U87 for example, will only pick up frequencies up to 20kHz. As you've admitted yourself, you can only hear up to 16kHz, but lets go with the more normal 20kHz as the example.

According to Nyquist's, aliasing would occur and be potentially audible with a sample rate of 40kHz or below. Beyond 40kHz the difference to the human ear would be inaudible. So, regardless of how well you wish to keep to the source material with a digital representation, it will be overkill. Since with CD rates we're already at 44.1kHz and covering up to 22.05kHz, what's the point? You don't write tunes involving bats do you? :D

I honestly do think that a lot of people got suckered into a marketing game where bigger=better. As yet I remain unconvinced but am willing for someone to offer me proof otherwise. :)
 
What using higher sample rates does give you, however, is the ability to manipulate the audio with higher levels of clarity and precision.

Like mentioned, it will be mastered to 44.1/16, but for the time the engineer works with the material, the response you get from working at higher sample rates makes it easier to work with.

Nyquists theorem of 44.1Khz is a minimum value in all honesty. Yes, the human ear cannot hear values above this, but that doesnt mean that they should not be captured in the first place. A lot of overtones/harmonics in these ultra high frequencies give an instrument tone, and so by sampling higher, you are able to manipulate a more 'correct' timbre of the instrument.

I find myself using 96k as much as possible. I've not used 196Khz, nor do i know anybody that does.
 
I'm sorry Lowe I disagree - I CAN hear the difference. When we did our blind tests a year or two ago in our main control room I admitt that getting 44.1/48kHz was guesswork but the leap to 96kHz was definitely there I don't care what anybody says...I heard it!

As far as I am concerned the higher the sample rate at tracking stage the more accurate the sound will be the original. Granted we master to 44.1 but thats mastering! Recordings have so much processing these days that I think it's important to maximise the digital accuracy at every stage to ensure a superior sounding mix/master.

+1 to Ozzie
 
Agreed - some instruments produce overtones and harmonics beyond that of the human ear. But like I said before - virtually no microphone picks these up! :) Even those which do go beyond 20kHz, the response isn't going to be even so you're potentially picking up an uneven response.

Food for thought. ;)
 
You must log in or register to reply here.
Back
Top Bottom