Opinion: The Mac Pro isn’t overpriced, it offers something nothing else does
flip.it
flip.it
I don't know why the author needs to spout bullshit about the specs like you can't build an equivalently (or higher) spec'd machine for less money. You just can't put MacOS on it without some effort and violating the ToS.
As for the MPX connectors, I don't see how those are different than the kinds of PCIE slots you'd see in top of the line ATX/EATX mobos.
As an aside, the improvement over the iMac Pro is largely proper cooling (Apple's AIOs haven't been able to take a fast multicore processor to respectable loads in like a decade) and availability of disk space worth a damn.
Yes - when you need a few thousand of these to render your SFX, you shop around for alternatives. Then some startup that recently built a cloud render farm for this exact use case will save you the CapEx, give you a manageable OpEx and prove this Op-Ed wrong. SFX studios are legendary for hiring and firing elastically, you think they care about Apple hardware lock-in?
Also, if you're a software business built on Apple, do some research - they'll fuck you over in a heartbeat. Maybe they'll block you on the Mac Store for using a private API [1], or retroactively apply a bogus rule [2] if they want to compete with you. Control your destiny by moving off an exclusive-Apple strategy
[1]: https://m.slashdot.org/story/362994
[2]: https://www.google.com/amp/s/www.theverge.com/platform/amp/2...
I guess the mac pro makes sense for animators, etc.
Is that because animation software is Mac only? Is there any hardware or software benefit unique to the Mac that they (animation software developers) rely on?
I could put a Mac Mini on my desk with half a dozen adapters sticking out of it, but even then I still can't upgrade the RAM. The last Mac Mini with user-upgradeable RAM was 2012.
The 2019 Mac Pro feels to me like an Apple Lisa. Even at 1/4 the price, it would seem steep to me. I want what it has, but there's no way I'll ever be able to justify that.
This is new territory for OSX-era Apple. It didn't used to be so expensive to buy an expandable Mac. You could get a PowerMac G5 for $1800, and a PowerMac G4 from $1500.
The 2018 Mac Mini is user-upgradeable up to 64GB. Storage is soldered on however.
We have already established [1] that even ColdPlay level musicians or Hollywood level Audio effects does not Mac Pro level computational power. Something we assumed for a long time I guess we now have evidence to call this a fact.
And everything in that video could have been done on a iMac Pro or even latest MacBook Pro.
So apart from SFX or Data Science / AI, I cant think of anything that requires these level of Computation power. And none of those are on the Mac Platform.
Putting multiple graphics cards in a machine has been common for a long time in gaming rigs. Crypto-miners run +12 graphics cards in an attempt to be profitable.
There's nothing unique mentioned in the article except the proprietary video stream processing, and I guarantee you, there's other high end machines that do the same thing if not more.
I really wonder what the author's specialty is, since he thinks these specs are actually unique to Apple.In fact, the more demanding your requirements, the more likely you need all of these fairly exotic components to work seamlessly together.
If you’re working on the next Star Wars movie, there are billions of dollars at stake; you can’t rely on some Linux box a guy you know put together, no matter how good the specs supposedly are.
Even if you could get comparable components, you’re not going to have the driver support you need. If you have downtime, you can’t wait for some guy on an email list who’s day job is not supporting this mission critical driver, to get back to you.
And you’re not going to get the performance of Apple’s APIs like Metal2, etc. which blow OpenGL, etc. away.
No, there's nothing like the current Mac Pro.
A quick look doesn't show a Dell Precision desktop workstation that goes beyond 48GB RAM, for example, while the Mac Pro goes to 1.5 TB.
I also didn't see anything like the MPX Module:
The MPX Module starts with an industry-standard PCI Express connector. Then, for the first time in a graphics card, more PCIe lanes integrate Thunderbolt and additional power provided to increase capability. With up to 500 watts, the MPX Module has power capacity equivalent to that of the entire previous-generation Mac Pro.
Didn't see a way to have up to 8K streams going at the same time or the ability to attach up to 12 4K displays on the Dell site.
It goes on and on. You can approximate a lot of what Apple is offering, but you can't duplicate what they're offering at a lower price.
There's so much Apple-designed silicon in this box, it's literally impossible for Dell, HP or anyone else to compete with off-the-shelf, commodity components.
For the target market Apple is going for, the cost of even a tricked-out Mac Pro is a rounding error.
Up to 3 TB RAM, 4 GPUs with 4 DP ports each if you want, dual processor if you want (so 56 cores), ...
Costs more or less the same as the Mac though, or more if you go higher-spec. MPX might be interesting, although I notice the Apple marketing comparing it to previous Macs and not competing PC-workstation cards, and of course application-dependent.
They were saying this one card has more processing power than the last generation Mac Pro.
Sure, the hardware is comparable if you don't take Apple's custom silicon into account, which by definition doesn't exist anywhere else.
None of the highest-end graphics cards--AMD Radeon Pro Vega II (64 compute units, 4096 stream units, 14.1 teraflops single precision or 28.3 teraflops double-precision) or the same card in the Duo configuration with two Vega II GPUs each with the 64 compute units, 4096 stream units, etc. are available from Dell.
Windows 10 is a big improvement over previous versions and the rollout of macOS Catalina was rough, but as things settle down, I believe Apple has the edge when it comes to ease of use and robustness.
And there's software like Final Cut Pro, which is an industry-standard in the movie business that only runs on macOS and is updated to take advantage of the new Mac Pro.
What good is testing when you release defective product anyway?
For the work loads that these will be GPU driven not CPU. Plus more CPU cores doesn't mean more performance, as software has to be optimized to handle so many cores and more cores means lower clock per core.
That being said, I do like AMD over Intel.
Statistics seem to agree till q3 2019. But don't know exactly why it wasn't after that.
A: save money and try to let a product die (mac mini, mbp, ipod)
B: invest to signal hope that a product is in fact not dead (mac pro, also mbp)
both approaches will eventually lead to an iphone-ipad-only apple because in case A things just fade away with the occasional life sign as to not lose too many customers.
case B, on the other hand, gives people what they were yearning for for years and keeps „professionals“ and press stimulated but at a ridiculous price point so that, b/c of low sales, the product line can finally be killed.
The (wholly understandable) frustration with Mac Pro pricing seems mostly due (1) to the continuing lack of lower-cost, non-Xeon modular systems in Apple's lineup, and (2) sporadic upgrades leading to eventually overpriced systems (e.g., the 2013 Mac Pro ca. 2019).
However, I agree with you that their business acumen is lacking, pricing at 25-50% over their best quality competitor would have vastly expanded the market, and their bottom line. But someone is advising them to be the Hermes/Versace of computers and somehow making business sense to decision makers, so this is what we get.
Indeed the market is made of all people, each with their personal and subjective assessment of value based on their own personal/subjective needs, criteria etc. What a product is worth TO YOU is your decision alone… and totally subjective, and… statistically irrelevant to the market at large.
The objective value measure though that matters to the market at large is: for what price can you be certain to find a buyer (in other words: "what the market will bear").