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.
Apple Mulling Preview of New Modular Mac Pro at WWDC in June
- Thread starter MacRumors
- Start date
- Sort by reaction score
It's ridiculous how long it takes... By the time the new Mac Pro is available for purchase, all rational users may have left the Mac platform long ago.
<sarcasm>Yay. Can't wait for June. I'm sure it's going to be exactly what people have been waiting for all these years, it's not going to be overpriced at all just like the iMac Pro and it will ship in 2019. </sarcasm>
I should really start looking into building a Hackintosh.
I should really start looking into building a Hackintosh.
Last edited:
Add ‘expandable’ to that too, which is more important.
What does beefy mean? I think you are well aware.
What does beefy mean? I think you are well aware.
If they {bleeped} up yet again, then RIP the bandaid off and get it over with - - because bad news does not become better with age.
They should go look in a mirror instead, because this entire mess is 100.00% a Leadership failure.
Broadly speaking, I'll take exception to most of these pricing decisions.
The basic reason is because we're buying technology, not lumps of coal & iron, and Tech has had a long history of price declines over time.
For example, SSD costs have declined by 75% since 2012 and we're now on the cusp of $100 per TB. That's why the 2018 MBP didn't need a price increase.
For a similar but much longer scale example, way back in 2006, I bought a 60GB "digital wallet" for $478, because that alternative was substantially less expensive than buying 64GB memory cards back then. Thirteen years later to the month, I bought some 64GB cards for $13.29 ... that's a 97% cost reduction, even before noting that (a) the old purchase was an alternative to an even more expensive option, and (b) today's memory cards have much higher Read/Write performance than anything available back then. If you want to time-average that, figure -7.5%/year.
Now apply this to the 2014 vs 2018 mini ... four years of reduced component costs .. and they actually had the audacity to ask for a price increase? Even if there's only a 50% applicability for component cost price reductions, they should have still generated over a $100 cost decrease. Did any of this go into the customer's pocket? Nope.
FWIW, I also did a head-to-head cost comparison for insurance purposes a few years ago on the Mac Pro vs Windows Tower PC ... Granted, this personal anecdote is workflow specific, but my bottom line was that Apple's iron was 36% more expensive: in round numbers, PC @ $5500/seat vs Mac @ $7500/seat (before Monitors). OS X is nice, but is it an additional +$2K/seat nice?
[doublepost=1550757862][/doublepost]
In counterpoint, when a $5K tool does the same job as a $10K one, I'll drop the bucks for the $5K and pocket the difference.
Case in point ... looking at buying some IR cameras right now and found that I can either buy a "does it all" for $50K, or buying a $10K & $12K camera which combined meets the same key requirements plus give me more flexibility. Getting the order started for $22K and banking the savings.
Last edited:
If they want to make a machine with applicability for VR, VR is basically a 4K Gaming workload, and AMD doesn't make a GPU that's any good for that.
Since WWDC 2017, this is the sum total on non-game VR titles for Mac. Three apps.
For Windows:
202 non-game VR apps.
Apple, don't for one freaking second claim you're interested in, or building for VR until you pull your fingers out, and deliver the Nvidia solutions, and user-upgradable retail GPUs that VR depends on. And if Metal can't compete with CUDA for GPGPU computation if Nvidia makes CUDA is available, then sod Metal, it's weaksauce clownshoes that deserves to be ignored.
Vinyl Reality was one of the titles Apple previewed at their "iMac Pro is good for VR" event just after that machine was released - it still isn't out for Mac. Pathetic.
Or hell, you could deliver a laptop with non-joke graphics - the windows world has laptops with the equivalent of double the top of the range iMac Pro when it comes to GPU power for generating immersive 3D environment There's dozens of laptops with full high quality support for VR (all of them Nvidia powered), Apple doesn't have a single portable system that can do VR without an eGPU. Again, Pathetic.
For a content creator working in VR toolsets, you'd better believe that "gaming" laptop as the pet Mac blogaratti like to dismiss it, is a smegton more professional than any "Pro" MacBook.
Since WWDC 2017, this is the sum total on non-game VR titles for Mac. Three apps.
For Windows:
202 non-game VR apps.
Apple, don't for one freaking second claim you're interested in, or building for VR until you pull your fingers out, and deliver the Nvidia solutions, and user-upgradable retail GPUs that VR depends on. And if Metal can't compete with CUDA for GPGPU computation if Nvidia makes CUDA is available, then sod Metal, it's weaksauce clownshoes that deserves to be ignored.
Vinyl Reality was one of the titles Apple previewed at their "iMac Pro is good for VR" event just after that machine was released - it still isn't out for Mac. Pathetic.
Or hell, you could deliver a laptop with non-joke graphics - the windows world has laptops with the equivalent of double the top of the range iMac Pro when it comes to GPU power for generating immersive 3D environment There's dozens of laptops with full high quality support for VR (all of them Nvidia powered), Apple doesn't have a single portable system that can do VR without an eGPU. Again, Pathetic.
For a content creator working in VR toolsets, you'd better believe that "gaming" laptop as the pet Mac blogaratti like to dismiss it, is a smegton more professional than any "Pro" MacBook.
Also, not every "pro" is self-employed and has the luxury of being able to order what they want, and may have to go through the common or lesser spotted procurement manager (with their distinctive cry of "I can't sign this off until you can explain why this $500 Dell won't do - PS. I'm now on leave for three weeks"). In a corporate environment, "equipment" and "human resources" are different budgets, managed by different people and suggesting that you might be able to amortise one against the other is dangerous, subversive thinking.
Puh-lease(rolls eyes)... not all of us work in a limited workflow. If your job involved one or 2 apps - perhaps.
So at the moment - I have open illustrator, Photoshop, Cinema4d, After effects. Premiere, Substance designer and painter - as well as Numbers / Evernote / Mail. Constantly bouncing between them is fantastic in OSX - horrible in windows.
Multi Application memory management just isn't as good on Windows - Just swapping around on Windows is not nice. My PC workstations WILL crash a lot. My Mac's don't... I do have a fully spaced up Hackintosh too and that does crash. A LOT! But that's a dark rabbit hole I don't really want to go down.
All of these live in their own space with independent macros and keyboard shortcuts as they are system level for the most part in OSX - so you can assign custom shortcuts to run multiple things in different apps at the same time - try that in windows.
Oh and very useful Mac only apps that don't have a Windows counterpart - well of course some of these exist in windows but none are as integrated or well designed.
Default folderX
SwitchresX
Keyboard Maestro
BetterSnapTool
Automator
I've had probably 15 workstations since 1996. And Mac have often been slower but Barr some exceptions (g4 cube) they have been way more reliable in day to day use.
Component costs for that 2014 mini were likely around $300, and they are probably similar or even higher now. So there’s no way $225 cheaper is even remotely possible. And just the loss of purchasing power of the US dollar accounts for virtually the entire price increase of that 2018 mini.
Apple’s gross margins haven’t changed in years. But looking at the 2014 vs. 2018 mini, I think you’ll see that $50 buys a lot, including:
—a 3.6GHz 65W desktop-class 8th generation quad core CPU vs. a 1.4GHz 15W mobile 4th gen dual core;
—four Thunderbolt 3/USB-C ports vs. two Thunderbolt 2
There are other upgrades like user upgradable RAM that maxes out at 64GB instead of 16GB, Bluetooth 5.0, USB 3.1 Gen 2 (10Gbps) and HDMI 2.0. Also, the mini now has Apple’s T2 chip. That’s technology making computing better/cheaper, but component prices aren’t necessarily going down.
re: your Mac vs PC pricing analysis, it is incomplete. Lower support costs, better resale value and other factors offset higher initial hardware costs for Macs. If you look at total cost of ownership, Macs are actually cheaper. IBM did a TCO study that makes for good reading.
https://www.cio.com/article/3133945/hardware/ibm-says-macs-save-up-to-543-per-user.html
Two years later, IBM has 135,000 Macs, 40,000 iPads and 275,000 Apple devices overall, including iPhones. And they deploy and manage it all with 78 support staff.
Last edited:
I frequently work in multiple heavy programs in Windows with no issues. Can you provide a citation for magical Multi Application memory management...I am not seeing anything on google
Last edited:
That is a weird turn of phrase. Memory management in Windows is fine. There are plenty of real problems with Windows, why make up stuff? Forced upgrades that frequently break things comes to mind. I'm tired of supporting family and friends when Microsoft chooses to push a button in Redmond and blow up their machines. Maybe the original poster used a Windows machine with 1GB memory or something.
This is not normal PC behaviour and you have either pulled this out of your rear end, or, you were using a computer that was ill designed for the workload you were putting on it.
Windows memory management isn't going to swap to disk and decrease performance (Which also happens in OSx) if there is sufficient memory.
this very complain of yours seems like you're comparing different class of machines an claiming that the more expensive better specced machine is better... Well, DUH.
you can't take a $3000 iMac fully loaded with 32gb of RAM and SSD, and compare it to a $300 computer with 4gb of RAM and spinning rust, and say that all things being equal, PC's suck in comparison to Apple.
Fundamentally, Apple computers are no different than windows. The way memory is handled is essentially the same to the end user. Neither OS are going to arbitrarily close down background running programs.
I have often hear similar complaints, and almost 100% of the time, the reason for the performance differences is the similar thing. they have either done something themselves to their computer to cause performance issues (or their IT, we aren't immune and often make user experience worse for our own purposes), or that the computer used is ill-specced for the job it's required to do.
and yes, I have very VERY much experience OSx having memory problems, that result in a spinning beach ball for seconds (minutes) while you wait for the programs working set being swapped from disk to memory. Again, this behaviour is quite universal for computing these days and is true of Windows, OSx and Linux based systems
Well, ignoring childish personal insults and the vast amount of incorrect assumtions you have made. A lot of windows apps and systems still use 32 bit code because Microsoft don’t want to annoy their enterprise customers which is insane - that and the clusterf that is the windows registry.
As an aside I used to be a MSE in another life. I know how awful windows is. I’d use Linux if it actually had the software I use.
Last edited:
you can claim anything, but the very comment I quoted of yours outlines a gap in your knowledge of understanding in how those processes work.I'm sorry if stating such feels like a personal attack. it was not meant as such.
FYI, an editorial correction - revision control fail on my part: I'd intended this to say that even if only 50% of the end product's cost was attributable to component price declines, this -$225 would have been only half of this (-$100).
And yet despite currency fluctuation headwinds, key prices in the USA have still come down, and significantly so.
Take the mini's 128GB SSD: the retail price of a non-Apple-proprietary 128GB M.2 (NVMe) has dropped from $75 in 2014 to $25 today ... that's a 66% decline which in principle is a $50 cost reduction (retail) from literally but just one component.
Sure, but to do the comparison, one really needs to find out how much Intel was charging back in 2014 for that 1.4GHz CPU and compare that to how much they're charging today for the 8th gen CPU.
Pragmatically, Thunderbolt is an Apple proprietary variant on USB-C which makes direct comparisons complicated. From a cost analysis perspective, I'd probably start with the "state of the shelf" costs for USB-C 3.2, even though I know that Apple has higher performance.
And again, an Apple decision to add a proprietary product which increases costs, which merits an assessment to see what the actual benefit to customer is, and how much it is worth. Given how we know that it also makes some of the customer's lifecycle management a headache, I'm not necessarily convinced that this nets out a positive.
Overall, I'm not disputing that the technology is advancing - - but I'm actually noting that these same advances are also being made in net costs. I've already provided the SSD example. Similarly, I was on one project two years ago to improve process yield at a chip Foundry, we successfully cut the cost of manufacturing by 80% (from $42/chip to $8/chip). This is why Tech doesn't follow conventional Inflation indices.
Oh, I know. The exercise was a straightforward insurance replacement cost question, which doesn't get into support/maintenance costs at all.
But even using your reference, its TCO savings claim is ($273 to $543) over a four-year lifespan, which even if I were to use the most optimistic end of this range still puts my Macs in the hole ... by roughly $1500/seat ... even after this sort of TCO is included. Again, this is a YMMV because I'm applying my own workflow use case, but the point still holds that when a Mac costs significantly more upfront than their comparable PC, even the TCO lifecycle management cost advantage isn't necessarily big enough to save Apple.
Case in point ... a did a very quick/lazy look at Intel NUCs as a mini alternative: at first blush, our $799 mini can be replicated for $546. Biggest caveat on this price is that we already have a site license for Win10, so I don't need to pay for a separate OS for +$150.
Editorial add: this difference of $243 between the NUC and mini would of course also be offset by OS-based TCO savings, although since this is a low end machine it would be akin to IBM's bottom range of $273 over four years, so its pretty much a net zero wash in this particular exercise as-is. Upping the bare bones specs such as to a larger SSD, changes the numbers too: at 256GB SSD, it becomes $636 vs $999 (delta = $363); at 512GB, $701 vs $1199 (delta = $498), and 1TB, $771 vs $1599 (delta = $828). Point being that even with just the mini, it doesn't take much for Apple's purchase price to outpace Apple's TCO advantage to make the PC the winner. -hh
Last edited:
Which one of your cited software apps is currently running 32 bit software on
windows?
I hear what you’re saying, and I don’t mean to suggest that Apple couldn’t cut costs if that was a priority. Rather they seem to try to limit increases, while adding features. T2 is a good example; Apple is adding it to all their hardware, and it replaces other components they’d otherwise have to buy from another vendor. Having it in all Macs simplifies hardware and software integration and reduces associated development costs.
They didn’t have to add the T2 but they did; it’s the SSD controller; the Secure Enclave allows secure boot and holds crypto keys for FileVault; it does full speed AES encryption/decryption for internal and external disks; and does audio processing and allows Hey Siri. (On laptops it does image signal processing for the camera, stores TouchID fingerprint data, and provides hardware audio disconnect of the microphone when the lid is closed.)
Thunderbolt 3 (not Apple proprietary btw) is another example. It offers port flexibility, and a ton of bandwidth for supporting tech like eGPUs. Again, it’s being implemented across all Macs but they have to buy a TB3 controller from Intel. Like the T2, it isn’t cheap, I think it’s around $15-20.
Your chip cost reduction example is interesting, and obviously allowed them to increase profits. But you can be sure the foundry didn’t cut prices 80% to their customers, rather they probably spent most of the savings internally, on R&D, improving other parts of their process that would increase future profitability, updating equipment and so on. (And that’s a smart thing to do.) Customers no doubt saw some of that passed on to them, but no management team wants to cut their revenue 80%.
Point being, even if some component costs go down, others can go up. Like I mentioned before, component costs were maybe $300 in 2014, and my guess is that it’s a struggle to keep that under control. DRAM prices, and flash also until relatively recently, were higher. Intel has not cut CPU pricing, but the 15” MBP has a hexacore instead of a quad core. Again, gains in technology and computing power, but not a decrease to BOM cost.
Overall, Apple’s gross margins haven’t been increasing, so it’s not as if they’re being greedy and just padding profits. But it’s a popular opinion around here. Apple introduces a new premium iPhone tier and people start yelling about Apple going crazy with raising prices. In reality, that’s only compensating for lower, slowing unit sales.
The XR is priced at $749, exactly between the previous $699/799 iPhone 8/8 Plus; prices didn’t go up year over year. But yes, Apple is selling a more expensive product with more expensive components inside it. That’s only a problem for someone not satisfied with the same tech from the 8/8 Plus that’s in the XR, and feels entitled to Apple’s top of the line offering, which is a moving goalpost, at no increase in price.
When the new Mac Pro is introduced, you can be sure we’ll here no end of complaints from the Apple-hate crowd—99% of whom would never buy a Mac Pro. What matters is what those in the target market think. They vote with their dollars. I’m very excited to see what they’re going to bring. Hope springs eternal, I guess
Last edited:
windows 64 runs most 32 bit stuff. Also alot of new games are 64 bit only.
/
Oh, I can very well see their strategy, particularly how it amortizes that product's development costs across a much larger base.
Oh, I'm not disputing what the T2 does. My point is that these "good things" also have trade-offs which need to be considered as well. For example, if the customer has other means of providing data security, then from that customer's perspective, because the T2 makes recoveries from hardware crashes more difficult, repairs a bigger hassle, etc, then the design trade-off isn't so great from their use case perspective.
Oh, its not that I don't appreciate the bandwidth potential of Thunderbolt - my concern is that its yet another example of Apple cramming stuff together for the sake of cramming it together, which leads to trade-offs. Much of the whole Mac Pro design discussion for nearly the past decade ultimately come back to how TB is functionally incompatible with PCI architecture, which requires ugly kludges to combine data and video onto a single cable. Putting them together is a solution in search of a real problem (because aesthetics isn't a legitimate excuse when it has forced you to have ugly add-on boxes, each with their own signal & power cables).
FWIW, I see this aspect of TB as a parallel to how Apple chose to decrease the capability of the iPhones by dropping the analog mic port and forcing wired headphones onto the Lightning port. Some people will claim that this wasn't a loss in capabilities, but from an engineering design perspective it most definitely was, because the capability to recharge the phone while concurrently listening through a wired headset is no longer possible out-of-the-box: to restore that capability, the customer is forced to either buy a Bluetooth headset or a 3rd party $39 dongle
True, but the point here is that Tech isn't a slave to the Consumer Price Index (CPI) for inflation.
Well, everyone is out to make a buck, as well as to try to increase margins wherever they can get away with it. Where Apple chronically shoots themselves in their own foot is in choosing proprietary stuff when commodity class components exist and can be used just as well, while leveraging outside-of-Apple market forces to drive down costs. The wasn't-quite-a-standard-M.2 SSD in the Mac laptops had been an example here.
One needs to be a bit cautious in assessing gross margins, as there's methods to bias that metric. For example, since Apple has slowed Mac development and allowed so many models to become "stale", they've had intermediate savings in component prices, etc, as well as longer amortization of fixed costs (both development and manufacturing) which would traditionally show up as a higher gross margin. But if you increase your costs elsewhere which are counted against this product line as an expense, it decreases the indicated profits.
And AFAIC, there's evidence of this happening in at least two places in Apple's SEC filings.
One place is in that new "Services" line. There's a statement hidden in the fine print which basically says that hardware that uses these Services is (internally) assessed an expense fee. This increases the cost to build that piece of hardware and thus, lowers the indicated gross profit margin. Now didn't the Services line just report that they had a huge (68%-ish?) gross margin?
Another place is in Apple's R&D budget, which is much more obvious on their 10-Q form: they increased their internal R&D spending YoY by $495M (+14.5%) ... yes, $5M short of +$0.5B.
-hh[/quote][/quote]
Oh, I can very well see their strategy, particularly how it amortizes that product's development costs across a much larger base.
Oh, I'm not disputing what the T2 does. My point is that these "good things" also have trade-offs which need to be considered as well. For example, if the customer has other means of providing data security, then from that customer's perspective, because the T2 makes recoveries from hardware crashes more difficult, repairs a bigger hassle, etc, then the design trade-off isn't so great from their use case perspective.
Oh, its not that I don't appreciate the bandwidth potential of Thunderbolt - my concern is that its yet another example of Apple cramming stuff together for the sake of cramming it together, which leads to trade-offs. Much of the whole Mac Pro design discussion for nearly the past decade ultimately come back to how TB is functionally incompatible with PCI architecture, which requires ugly kludges to combine data and video onto a single cable. Putting them together is a solution in search of a real problem (because aesthetics isn't a legitimate excuse when it has forced you to have ugly add-on boxes, each with their own signal & power cables).
FWIW, I see this aspect of TB as a parallel to how Apple chose to decrease the capability of the iPhones by dropping the analog mic port and forcing wired headphones onto the Lightning port. Some people will claim that this wasn't a loss in capabilities, but from an engineering design perspective it most definitely was, because the capability to recharge the phone while concurrently listening through a wired headset is no longer possible out-of-the-box: to restore that capability, the customer is forced to either buy a Bluetooth headset or a 3rd party $39 dongle
True, but the point here is that Tech isn't a slave to the Consumer Price Index (CPI) for inflation.
Well, everyone is out to make a buck, as well as to try to increase margins wherever they can get away with it. Where Apple chronically shoots themselves in their own foot is in choosing proprietary stuff when commodity class components exist and can be used just as well, while leveraging outside-of-Apple market forces to drive down costs. The wasn't-quite-a-standard-M.2 SSD in the Mac laptops had been an example here.
One needs to be a bit cautious in assessing gross margins, as there's methods to bias that metric. For example, since Apple has slowed Mac development and allowed so many models to become "stale", they've had intermediate savings in component prices, etc, as well as longer amortization of fixed costs (both development and manufacturing) which would traditionally show up as a higher gross margin. But if you increase your costs elsewhere which are counted against this product line as an expense, it decreases the indicated profits.
And AFAIC, there's evidence of this happening in at least two places in Apple's SEC filings.
One place is in that new "Services" line. There's a statement hidden in the fine print which basically says that hardware that uses these Services is (internally) assessed an expense fee. This increases the cost to build that piece of hardware and thus, lowers the indicated gross profit margin. Now didn't the Services line just report that they had a huge (68%-ish?) gross margin?
Another place is in Apple's R&D budget, which is much more obvious on their 10-Q form: they increased their internal R&D spending YoY by $495M (+14.5%) ... yes, $5M short of +$0.5B.
-hh[/quote][/quote]
1) Thunderbolt tech is not incompatible in the least with PCIe architecture; it literally provides external access to the PCIe bus itself.
2) Proving an accusation of Apple of manipulating its books to decrease gross margin—which Wall Street hates for any computer hardware manufacturer—will need more evidence than “it’s possible”. If your allegation was that Apple was cooking the books to increase gross margin, it would be much more believable.
Part of auditing Apple’s financials would be their CPA scrutinizing methods of attributing expense across product categories. As long as the method is consistently applied, the financials are comparable from period to period. Apple would not be allowed to change the parameters of the cross category allocations just to manipulate quarterly results. Any changes would need to have a valid accounting principle behind it, and serve to increase the accuracy of the financials—and would be disclosed.
3) R&D is an operating expense and is not part of cost of sales; it’s below the line and has no effect on gross margin. Thus your claim that YoY increases in R&D expense are being used to drive down gross margin is completely invalid.
I really don’t have much interest in discussing your thoughts about Apple’s SEC filings or otherwise, so I’m going to bow out of this conversation, it having strayed far afield of the topic.
Last edited:
I had one of the 2013 Mac Pros. I loved how silent it ran. It was a good machine, and I really liked it. However, it was never upgradable other than the SSD and the RAM. I sold it after 2 years for about 80% of the original price I paid because I realized it was more power than I needed for what I was working on, and I wanted to recoup some money. I'd almost buy any new Mac Pro sight unseen because the resale value is just incredible.