Pro Tools on the new Mac Pro vs old 2007 Mac Pro

[Protootr]

Want to know how the new Mac Pro compares to an old Mac Pro 2007 running Pro Tools HD Native?

I have documented installing PT HD Native on a new Mac Pro and tested it with a CPU demanding session with over 300 plug-ins. Plus I compared it with Pro Tools running on my 2,1 Mac Pro 2007.

I also did a DVerb test and a VI test. If you want me to test anything or have any questions regarding the new Mac Pro running PT HD, just post it here and I'll give you the results.

http://www.protootr.com/pro-tools-on...nd-new-mac-pro

[lancemcv]

Nice comparisons.

[Protootr]

Thanks, glad you liked it!

[JFreak]

Originally Posted by Protootr

If you want me to test anything

Try to repeat this test I just made with my latest top-of-the-line MBP.

It is a 96kHz 32-bit surround session that has 16 tracks of 5.1 input, each with ReVibe and SSL Channel. Recorded 10 minutes of silence without a single hiccup. (was actually reading DUC posts during the test). It uses 96 voices and generates +20GB dummy data. Recorded on the laptop system drive against recommendations.

[LoZion]

Good job! 6 core is the way to go...

[Protootr]

Originally Posted by JFreak

Try to repeat this test I just made with my latest top-of-the-line MBP.

It is a 96kHz 32-bit surround session that has 16 tracks of 5.1 input, each with ReVibe and SSL Channel. Recorded 10 minutes of silence without a single hiccup. (was actually reading DUC posts during the test). It uses 96 voices and generates +20GB dummy data. Recorded on the laptop system drive against recommendations.

The Waves SSL 400 bundle is funny enough the only bundle I own from Waves.
But besides the Slate plug-ins they are the only plug-ins I haven't got running in 64-bit yet. I'd like to run this test for you but I haven't updated my WUP yet to version 9. Maybe I'll do that later today and do the test for you.

[Protootr]

Originally Posted by JFreak

Try to repeat this test I just made with my latest top-of-the-line MBP.

It is a 96kHz 32-bit surround session that has 16 tracks of 5.1 input, each with ReVibe and SSL Channel. Recorded 10 minutes of silence without a single hiccup. (was actually reading DUC posts during the test). It uses 96 voices and generates +20GB dummy data. Recorded on the laptop system drive against recommendations.

What was the setting for you Hardware Buffer Size?

[JFreak]

Just checked, and it was 128 -- was spiking using 64 buffer.

[Protootr]

Originally Posted by JFreak

Just checked, and it was 128 -- was spiking using 64 buffer.

Okay, I just had a spare 15 minutes to do your test but in my case with Revibe II and Avid's Channel Strip.

One thing that struck me, and which I also noticed during the other tests, is that at low buffer sizes the outer cores 1, 11 and 12 are hardly doing anything in Avid's perception. The meters ramp up to core 6 and then down again with 1, 11 and 12 hardly at work. I believe what Avid is showing us with their meters might not correctly reflect what is actually going on with the cpu.

Of course Pro Tools hasn't been approved yet for the new Mac Pro and some tweaking might have to be done. Your CPU shows a reasonable even balance accross the different cores albeit at a 128 sample buffer instead of 64 but still. So it might be the new CPU that needs to be supported.

I updated my blog post with my findings.

This were my results with your test JFreak:

[Protootr]

Oh, by the way. The other thing that is strange is that Pro Tools is saying it's only using 94 voices while there are definitely 16 tracks of 5.1 surround in use.

Have no explanation for that and it might also be a sign that some bugs need to be ironed out with the new Mac Pro.