The "I'll wait for Haswell Mac Pro" Thread

[ZnU]

With the iMac posting a HIGHER single core score, it makes me wonder how often we're 'feeling' the speed of that single core speed, and how often we're feeling the effects of multiple cores. In general 'use', surely the snappiness etc is more correlated to that single core frequency than multiple cores? I'm not sure how far developed Grand Central Dispatch is yet, so I'm making a lot of assumptions here.

It is very possible that if with respect to 'snappiness at idle' (i.e. how responsive the machine is when it has nothing to do but respond to user interaction) there will be little or no benefit for a Mac Pro over a high-end BTO iMac. The good news is an iMac is already pretty damn snappy these days. In fact, a lot of this lack of difference is a consequence of the iMac already being a pretty damn high-end machine, not of any lack of impressiveness on the Mac Pro's part. For instance, Apple could have held back PCIe flash as an exclusive for the Mac Pro, but instead they offer the option on the iMac and seem determined to make it standard on every machine they make, something which no other personal computer vendor is presently doing.

I expect most of the benefit of the 12-core I have on order won't come in the form of responsiveness to user events, but in the form of e.g. not dropping frames on shots with complex grades in Resolve during real time playback, and, to an even greater extent, will show up when I'm not even in the room... when I've set up a 15 hour encode or render and walked away.

 

[hugodrax]

Are the Haswell XEON's purported to provide any substantial gains in CPU performance? I know they will be touting power savings, but I'm not really interested in that. I'd actually like to see the GHz go up a bit

Yes. Pretty significant gains with software compiled to take advantage of AVX2 and FMA.

So when newer versions of OSX appear and newer software you will see the difference with any software that leverages the new haswell features.

Linpack shows close to a 70% jump in performance between an equally clocked ivy bridge and haswell when using the new avx2/fma/gather features.

 

[theSeb]

I guess it comes down to how much of a jump in performance I will honestly see?

With the iMac posting a HIGHER single core score, it makes me wonder how often we're 'feeling' the speed of that single core speed, and how often we're feeling the effects of multiple cores. In general 'use', surely the snappiness etc is more correlated to that single core frequency than multiple cores? I'm not sure how far developed Grand Central Dispatch is yet, so I'm making a lot of assumptions here.

Sure, Photoshop will feel 'smoother' using more beefy GPU's. And if Photoshop can make use of dual D300s, D500s and D700s, then that would be interesting.

I think I really need to take one of these nMPs for a spin at an Apple store with Photoshop running, and a bunch of other apps, with a couple of screens connected, before I can make a final judgement call.

I would not rely on Geekbench as any indicator of real world performance. I have long suspected that it does not take thermals into account. Tonight I have proven this.... on a rMBP. I'll prove this again when I receive my nMP.

 

[Quash]

I would not rely on Geekbench as any indicator of real world performance. I have long suspected that it does not take thermals into account. Tonight I have proven this.... on rMBP. I'll prove this again when I receive my nMP.

I think you're absolutely right. A rMBP/Mac Mini can run a great sprint but not a marathon. It will throttle and on top of that is noisy as hell if you put any kind of real stress on it.