Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Any single-threaded operations will be much faster on a 3.70GHz nMP than with a 3.46GHz cMP. The latter only has 80% of the single core performance of the newer system.

Hi Nigel,

Yeah I understand, the single core performance of my 4790k hackintosh is above that of any nMP I believe? Where did you find that 80% number? I'd be interested to see that information.

Interestingly, I spent yesterday evening testing my 4,1>5,1 against a guy in the studio next to me's quad nMP. There was very little noticeable difference in general use, loading programs, spotlight etc. The main difference we saw was heavy project opening times, the nMP was pulling up the same project over 10 seconds quicker than mine. Also, it was snappier loading any plugins with a crazy GUI - Omnisphere took 3 or 4 seconds more - although I'm not sure if the graphics card has more to do with that than anything else. The Logic Benchmark test was confirmed though, he could only hit 94 tracks and I was hitting 128 (strange how it changes?) - I believe Logic X + El Capitan really squeezes the life out of every thread.

Overall he was pretty devastated - he payed over £3k 6 months ago.
[doublepost=1468404777][/doublepost]
Any single-threaded operations will be much faster on a 3.70GHz nMP than with a 3.46GHz cMP. The latter only has 80% of the single core performance of the newer system.

Apologies, I read that as the single core performance was 80% faster. If I had to put a number on it I'd say that's pretty much how it felt. But in real world use I didn't feel it would affect work flow significantly, and I'd happily trade a few seconds of plugin loading every now and then for an extra 30+ tracks to work with.
[doublepost=1468405057][/doublepost]
get this.

http://www.ebay.com/itm/Apple-Mac-P...379058?hash=item3ac6fa0c32:g:76oAAOxykMpTIAJE

12 cores complete tower for EDIT $1650.

and then get a MvC 980ti...or wait 2 months for the new OS to see if it supports 1080/1070 or maybe AMD?

But this machine will be fly for days.

That's a great deal - but I'm in the UK. Thanks.
 
Hi Nigel,

Yeah I understand, the single core performance of my 4790k hackintosh is above that of any nMP I believe? Where did you find that 80% number? I'd be interested to see that information.

Intel Xeon X5690 @ 3.47GHz
Single Thread Rating 1523
http://www.cpubenchmark.net/cpu.php?cpu=Intel+Xeon+X5690+@+3.47GHz

Intel Xeon E5-1620 v2 @ 3.70GHz
Single Thread Rating: 1915
https://cpubenchmark.net/cpu.php?cpu=Intel+Xeon+E5-1620+v2+@+3.70GHz&id=2047

1523 is 79.53% of 1915 (1523/1915)
1915 is 25.7% greater than 1523 ((1915-1523)/1523
 
In the recording studio, we tend to care little about single thread performance because DAWs, in general, make very good use of multiple threads.

I have always figured the Geekbench multi-thread score correlates at least roughly to track count, but in this case, the 2013 4-core score of 14,320 is 89.5% of the 16,000 modded 2009, yet it can handle only 73.4% as many tracks.

Curious about a comparison using VIs... thanks for sharing, Mr. Tiger.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.