iMac test

[Rainman]

I've been meaning to do this for a long time and finally got around to it today. I commandeered my wife's iMac to see what it would do. Like many posters right now, I'm in the market for a new Mac, my current one is a Power PC G5 dual processor 2.0 GHz and also a 13" MacBook Pro Intel core 2 duo at 2.26GHz. So, not having any real idea what a current Mac Pro might be remotely like I decided to load DP 7.2 in to my wife's iMac. It's not a scientific test but in many ways, a real world test for me, I rarely run more than 3 VI's at a time and typically average 16-24 tracks/song. I mainly work with songwriters and bands. I had to import the tracks from my MacBook since I only installed DP for this test (and have uninstalled as well I might add). So, first the specs:

2009 iMac i7 Quad core @2.8GHz (iMac 11,1)
8GB ram
8 MB L3 cache
Turbo Boost
Hyper-Threading

The song had 8 stereo tracks and 26 mono tracks. I set up 6 aux tracks, 3 for verbs and 3 for sub mixes. I used a total of 4 Proverbs, 3 on the aux's and 1 in a channel strip. One master fader. 19 of the tracks were drum/percussion with lots of nice bleed and I put EQ/comps on just about everyone of them. As I said, I just used DP so all the channels with processing were the Pre-set strips, it's quick and useful for this test, just wanted to put a load on. I used a total of 46 plugs, 16 MW levellers, 11 MW Eq's, 1 delay, 1 plate, 1 trim and the rest were various and sundry. I had volume automation going pretty much continuously on 15 of the tracks, the song was 6:09 mins. in length.

Once everything was loaded up I set the playback buffer at 1024 which is what I'd have to do on my G5. The activity monitor barely showed up, I was quite surprised at this. I'd installed MenuMeters on the iMac (from Raging Menace, very handy) and watched all 8 of them (hyperthreading) as I did a playback, they coasted along, one or two occasionally hitting 50%. I dropped the buffer to 512 and still the activity monitor barely budged, perhaps at 1/8 of the screen. Continued to drop the buffer, 256 more activity, 128 about 1/4, and finally 64 where activity sat just under a half and spiked a bit just over half. Really quite impressive. MenuMeters indicated fairly steady activity at 46%-56% on the cores. I bounced the entire affair to a 2 track mix and it clocked in at 2:15.

Obviously, this model of iMac and it's 2010 counterpart are no slouches when it comes to audio. I know that many people work differently than this but I think this somewhat un-scientific test indicates that for many others, an iMac may be all they need. I'm tempted myself as the 27" screen is absolutely gorgeous, all 34 tracks fit nicely across it. Unfortunately, I really need the drive bays a Mac Pro has but it has me thinking that a Quad core Mac Pro is no slouch at 3.2 GHz.

Anyway, that's all

Rainman

[BKK-OZ]

...it has me thinking that a Quad core Mac Pro is no slouch at 3.2 GHz.
Anyway, that's all
Rainman

Good work Rainman, very insightful post, thank you for taking the time to do the work and write it up, its appreciated.

Unfortunately, sometime ago I talked myself out of an iMac (just not enough expandability), so while your post is very interesting, it doesn't help me.

So, now pls do a similar test with a first generation (1,1) Mac Pro Quad-core (2.6Ghz or 3Ghz, I'm not fussy) because these machines are what I am looking at via some eBay auctions. After you do that, pls also test the Mac Pro 8 Core (2.8ghz), because there is an auction for one of those on @ the moment too...

[spencer4t]

I actually bought nearly the same imac you mentioned in your post. I mostly work with VI's so I needed the speed. I cannot complain about the imac, it has only been pushed when I put an insane amount of VI's together just to see what it would do. I do not really need much expandability at this point so the imac has been wonderful. Just my own situation I guess