Hey guys - been looking through a few threads and can't seem to see any comparisons between the new 2.3 vs 2.4ghZ processors. There seems to be very little performance difference between the two - < 7%. I spend most of my time in FCPx and was wondering who else does and has paid the extra for the 2.4. Is there any other difference between these two CPUs apart from the small clock speed difference? That 7% might back a difference on h.264 exports - or should I just put the $200 towards an eGPU?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
2.3 vs 2.4gHz.
- Thread starter Craigy
- Start date
- Sort by reaction score
I didn't see any tests, but from looking at the specs... I seriously doubt that you will see any difference between the two i9 models. With the 2018, going for maxed out i9 could be somewhat justified by the higher burst performance and cache. With the 2019, the clock delta is 200Mhz at most (which is going to be unnoticeable in practice) and the cache size is identical. I just don't see any point in paying that premium for a CPU with almost identical spec.
"Faster CPUs and/or multiple cores will render faster. Faster GPUs may render faster, depending upon the codec; ProRes takes advantage of the GPU. However, faster CPUs or GPUs don’t achieve greater quality, just greater speed."
Source: https://larryjordan.com/articles/final-cut-pro-x-can-you-render-faster/
"UPDATE. As the processors and speeds have changed, read this now as: Editing does not require as fast a processor as video compression. The base level processor will be fine for most editing, while a faster processor will benefit compression."
Source: https://larryjordan.com/articles/thoughts-on-picking-the-best-computer-for-video-editing-video/
Given that the multicore performance is heat-constrained, and not keeping up with the peak power draw/frequencies, multicore applications stand basically zero chance of being faster on the 2.4 vs 2.3. Someone did bring up that the HK version means a slightly different cooling potential, but I'll have to see that to believe. However, I'd be interested to see how the 2.4, with the 5.0GHz peak boost does on single threaded tasks compared to the 4.8 max turbo on the 2.3. Theoretically, with only a single core pumping away, thermal issues will not come into play and we'll just get the raw capabilities of the processors.
So the cooling architecture on the 2.4 is different than the 2.3? If that is the case with a potential cooling gain then that would translate into less throttling I guess - so more performance. I'm mostly on FCPx / Logic / Adobe CC. I am not starting to shoot more 10 bit / 4k footage - the price increase doesn't bother me in buying the 2.4 over the 2.3, just didn't want to 'waste' the money on the 2.4 is the difference was so negligible. But if the 2.4 cools differently to the 2.3 then that may swing it for me.
There are no differences between the 9880H and the 9980HK when it comes to cooling. Mobile CPUs do not have a lid on top with paste or a soldered heat spreader like desktop CPUs have. You cannot de-lid a mobile CPU. The CPU is entirely dependent on the cooling solution the OEM provides. The HK is simply an unlocked CPU as it does allow overclocking, just like the i9-8950HK. However, as with all unlocked CPUs, overclocking is not always a good idea.
See this picture - https://d3nevzfk7ii3be.cloudfront.net/igi/LWMIrFZIcqS4bOZU.huge
The CPU is in red, it only has the necessary materials for the OEM to attach it to the logic board, but no lid, like the 9900K has, which uses a soldered IHS.
You will not get better cooling with the i9-9980HK than the other two CPUs.
https://www.anandtech.com/show/1425...l-the-desktop-and-mobile-45w-cpus-announced/2
So - in reality unless you want the 6% performance increase then stick with the 9880H 2.3 - And I'd be much better off putting the $200 USD towards an eGPU for when I'm using Resolve etc?
In a separate thread, I am having a more esoteric conversation around this very topic. My succinct answer is yes, because Resolve scales up with multiple GPUs and VEGA 56 and 64 GPUs have dropped in price to where they are very cost effective options. If Apple gives us a Mac Pro preview at WWDC, and disclose GPU details, that opens up the door for some other officially supported options (Radeon VII, Navi 7nm...maybe NVIDIA...maybe). You can find several Youtube videos that demonstrate eGPU scaling in Resolve and FCPX and the gains seem more tangible than a 4-7% CPU gain will. But ultimately the decision is yours...$200 is not that much money and then you never feel like you left something on the table, but I would pass on it myself and bulk up on GPU horsepower since it sounds like your workflow would benefit more from that. Good luck!!!