Upgrade or expansion? Looking at what the slots on the Mac Pro were used for primarily, (GPU, ProRes acceleration) and those are built into the CPU now, what expandability would they still need slots for?
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.
MP 7,1 The Next and Final Intel Mac Pro
- Thread starter strangerthanlight
- Start date
- Sort by reaction score
Might be the difference in video cards. I think the Navi card is more optimized for video stuff than Vega.
On this page:
Do Adobe apps work on Mac computers with Apple silicon?
Adobe is actively working to build apps that run natively on Mac computers with Apple silicon. Many of our existing apps can run on these devices using Apple's Rosetta 2 technology.
Personally, I’d call the 3 seconds a wash. If I can start a process, leave my desk and come back when the fast one would have been complete, and only wait 3 more seconds for it to finish, that’s well within my personal range of “dealing with”.
I’d say, if you don’t need what’s essentially the same performance of a Mac Pro in a portable form factor for Apple Silicon compiled apps, AND still make heavy use of apps that aren’t compiled for Apple Silicon, the current MBP’s are likely not going to be a fit for you.
Oh, man. I've been laboring under a misapprehension. I thought the latest Adobe releases that brought (most of) their software into running Apple silicon natively included After Effects. Sure enough, that one's just in beta.
And you're right on the export/re-encode differences being nearly a wash. I think I should run some more tests this weekend with some longer projects (and subsequently longer export times), to see if the gap widens or shrinks significantly. And maybe see about downloading the After Effects beta. I'll share results here later, just in case anyone's interested.
Apple's return window for orders arriving after Nov 1 is extended through Jan 8, so I have some time to make sure.
I see myself buying a closed system, but I cannot see myself spending $5K+ on a machine where I can't upgrade some components to expand its power or adjust its flexibility as needed. Unless marginal but absolute speed is the overriding consideration above all else, the economics of buying a cheaper machine and upgrading more often versus a giant expensive rig just don't make sense. Hence why I got a Mac mini instead of the 7,1—came with legacy software support, and while it's much slower in many respects I could get 2+ replacement machines in next few years, and still come out ahead.
Expansion for SSDs, extra GPUs (provided Apple can be bothered to release discreet ones)
Else the cylinder could have continued to come with new versions every 3 years and we needn’t have been asking for expansion
Yeah, I’m thinking that whatever’s next is going to be a lot like the cylinder BUT without the slots… so I guess more like the Mini.
Perhaps they’d allow storage updates, but my assumption is that they’d provide for storage externally via Thunderbolt.
Well storage can exist outside the system but I would like to see discreet GPU Modules from Apple.
The holy grail of CG is real-time but the hardware doesn’t keep up once you go beyond a certain data threshold.
Apple has already demonstrated you can enable significant amounts of memory access to GPU to work on big scenes which were once the sole domain of CPUs.
A discreet one could add extra grunt without throwing away the entire system (or have to buy a whole new one just for the GPU)
Apple could even release these modules as eGPUs for non Mac pro systems.
Last edited:
Apple’s current architecture calls for the GPU to access the same memory as the CPU. Move that GPU away from the CPU and onto an external bus (card slot or eGPU) and your performance drops precipitously. The future will be known for certain soon, but I’m still thinking that the only GPU in a Mac going forward will be an integrated GPU.
Hmm. Most likely.
Anyway I was hoping for a discreet GPU along with the SOC one. That way the base system performance doesn’t get hit. For all live tasks, the user will be working off the main chip GPU..for other ones where user interaction isn’t required… a discreet variety (with more cores..even a dual GPU version ) would be very powerful.
That’s how I and users who have multi GPU systems use the setup.
Have one GPU for display/interactive work and the others for pure compute tasks.
Last edited:
Bingo, I don't care if a discreet GPU has lower video export or compute performance or whatever, literally couldn't care less about it. What I can't live with, is the number of screens, or the scale of a 3D viewport and its performance that I can drive, being tethered to buying a whole new computer, unless that computer costs as little as a GPU.
I love my 7,1 because of it's design and expansion for tinkering and enthusiast/hobby use. If for commercial use, there is no need for expansion. Every few years tech has substantially improved, upgrading the old system is still slower than the latest offerings or more expensive.
In any professional usage, which usually means system is worked 24/7, existing components are due for replacement. It's more cost effective to just replace the complete system, then partial replacement.
It's gone from old days in 80s and 90s where personal computing were mainly towers and parts were expensive and less abundant. An upgrade also requires time and risk.
Mac Pro's upgrades are expensive. The GPU upgrade cost as much as a Maxed out M1 Max.
Everyone seems more focused on the benchmark comparisons. I think we all got suckered into marketing and youtube videos, that a few seconds in improvement, we must buy the latest.
No one works that fast or efficiently. In reality, most work time are wasted in distractions ie. tapping at phones, breaks, spacing out, easily wasting minutes and hours of a day.
Unless is mission critical industries, where a millisecond lost is millions of dollars lost in transaction or life and safety situations.
In any professional usage, which usually means system is worked 24/7, existing components are due for replacement. It's more cost effective to just replace the complete system, then partial replacement.
It's gone from old days in 80s and 90s where personal computing were mainly towers and parts were expensive and less abundant. An upgrade also requires time and risk.
Mac Pro's upgrades are expensive. The GPU upgrade cost as much as a Maxed out M1 Max.
Everyone seems more focused on the benchmark comparisons. I think we all got suckered into marketing and youtube videos, that a few seconds in improvement, we must buy the latest.
No one works that fast or efficiently. In reality, most work time are wasted in distractions ie. tapping at phones, breaks, spacing out, easily wasting minutes and hours of a day.
Unless is mission critical industries, where a millisecond lost is millions of dollars lost in transaction or life and safety situations.
Interesting
How is the interface experience in After Effects on the Mac Pro? I just sent back my Mac Pro to Apple because of the incredibly sluggish experience of AE. I mean, on a normally heavy project (wich runs fine on my old 5,1) it's just unusable, spinning wheel at every single click (even no click at all)
I tried different AE versions, from 17.5 to 2022. All the same
I tried different monitors, plugged in TB, HDMI...
Big Sur or Monterey, no change
The only thing that worked fine was the render, super fast with Muti Frame enabled
Some other people here seem to have the same experience, but I feel it's not everyone, so I'm curious
Oh, if that’s the case, then wasn’t the GPU being used for compute always a backup for the lack of available specific compute units? Now that there ARE specific compute units available, that means less of a need for GPU’s. Although I COULD see a proprietary slot for offloading to a proprietary compute expansion card, just like Apple did with AfterBurner since they don’t need AfterBurner anymore as those functions are built-in.
If it didn’t have it’s own RAM, though, any compute done there would be a LOT slower than on-chip. Maybe a daughter card for another Apple Silicon system
I tested a maxed out M1 Max accross various video/3D type apps, vs a maxed out Mac Pro 28 Core, afterburner, and Quad W6800x MPX. (two duos)
The M1 Max beat the Mac Pro in anything pro res, which is impressive. But with r3d raw and other GPU reliant codecs, and 3D work, the quad W6800x still were a good deal ahead.
But...if you're comparing a non-maxed out Mac Pro, like a 16 core with maybe just a single W5700x or a single Vega II, I'd likely favor the M1 Max even with the r3d raw stuff. It's really good and nipping at the heels of the Mac Pro.
I have no doubt the Apple silicon Mac Pro will be amazing, and either match or beat the current maxed out Mac Pro without issue. (But that's still at least a year or more away)
Will Apple update the Intel Mac Pro? I am not sure, I would hope so, but exactly with what CPU updates is the question. Otherwise the only other MPX option I could see is maybe a W6900x Duo at the top end. (But will people really pay $10-12k for that GPU with the Apple silicon Mac Pro coming at that time? I'd still rather get the cheaper W6800x Duos..
The M1 Max beat the Mac Pro in anything pro res, which is impressive. But with r3d raw and other GPU reliant codecs, and 3D work, the quad W6800x still were a good deal ahead.
But...if you're comparing a non-maxed out Mac Pro, like a 16 core with maybe just a single W5700x or a single Vega II, I'd likely favor the M1 Max even with the r3d raw stuff. It's really good and nipping at the heels of the Mac Pro.
I have no doubt the Apple silicon Mac Pro will be amazing, and either match or beat the current maxed out Mac Pro without issue. (But that's still at least a year or more away)
Will Apple update the Intel Mac Pro? I am not sure, I would hope so, but exactly with what CPU updates is the question. Otherwise the only other MPX option I could see is maybe a W6900x Duo at the top end. (But will people really pay $10-12k for that GPU with the Apple silicon Mac Pro coming at that time? I'd still rather get the cheaper W6800x Duos..
When they introduced the M1 they said it was the first Apple Silicon in a transition that would take two years.
When they introduced the M1 Pro and M1 Max they said they were now 1 year in their transition.
So in around 12 months the entire line up should have transitioned exclusively to Apple Silicon.
Some surmise Apple isn’t going to release an intel Mac Pro next year. In which case, the 6900x duo also won’t make sense, not when you consider the power envelope of such a module.
However by not releasing an updated intel tower, Apple would have killed 3 desktop pro machines after just one generation. A terrible record to have.
And all of that started when Apple jumped the gun with the cylinder. I could go and on about the ramifications of that - and there really is no other proper way to put it - stupid move, but much water has flown under the bridge since 2013 and there is no point reiterating the topic.
When you consider that over 80% of the Macs folks buy from Apple are mobiles and another 80% of those that are left are iMacs, killing 3 desktop pro machines after 1 generation makes sense. These are systems that sell in the tens of thousands to hundreds of thousands, tops.
While at the bottom and middle, folks are still doing pretty much the same things, at the top end, market conditions and what those few thousand folks want could change dramatically over a few years. With a buying group that’s sooo small, it makes sense to make what that small group wants rather than iterate on something that doesn’t suit their need anymore.
Very few of that soooo small group would have asked for a dead end cylinder or an iMac style as a Mac Pro tower replacement. Hence you got the 2019 Mac Pro.
Apple could well have pivoted fast after the cylinder if that small group’s concerns were important to Apple. Instead we got radio silence for years and ..an iMac (with the same dead end features as the Cylinder )
They probably didn’t have to pivot because for all it’s shortcomings, that small group couldn’t purchase a faster Mac than the cylinder Mac Pro even if they wanted to. Apple expected, going into it, that the sales of the Mac Pro would be SO small, that a factory in the US would be able to handle the load. SO, if their goals were a tiny amount of sales, and they saw a tiny amount of sales, then it met it’s goals.
Any future Mac Pro will likely see similarly tiny sales and, based on recent history, a complete redesign to meet current workloads. And it will sell in limited numbers for 3-5 years when they either redesign again, OR sales have dropped to the point where it doesn’t even make sense to make it anymore.
So why make the 2019 Mac Pro or even the iMac Pro if the cylinder met the company’s goals ?
Are you suggesting Apple wanted/wants out of the Mac Pro segment ? Then why not do it now (or when they dragged their feet with the cylinder) ?
Last edited:
I placed an order on a m1 Max Mac Book pro, seems interesting to say the least. Im curious of how it will manage large 3D files I use for work, seems like a fantastic tool for people who work with video only.
Will post my findings later.
Will post my findings later.
That’s what I’m saying, those systems met the goals of both what Apple wanted to sell and what those users wanted. When those wants (Apple’s and the users) changed, Apple had no problem dropping and replacing it with something new instead of iterating the old thing.
I think Apple wanted/wants to move the Mac Pro price/features upward to where they are now, effectively reducing the Mac Pro segment over time and allowing them to focus on, make, and sell the mobile devices that the vast majority of their customers want. If they eventually get to a point where the Mac Pro is starting at $10,000 and selling only in the thousands per year, I’m sure they’d be fine with that.