I've nothing against Apple targeting the pro market, even though its current hardware offerings are anything but pro, but I really wanted a prosumer desktop Mac to replace my cheese grater.
Why wouldn't you buy a 5K iMac or iMac Pro? You're not going to have any prosumer software that won't run significantly better on those machines than a 2012 or earlier cheese grater. I bought the low-end standard 5K iMac and it blows away my old 2009 8-core Xeon Mac Pro.
Because they have a built-in monitor. The Mac Mini is too small and not powerful enough. The form factor of the Pro cylinder is perfect for the prosumer market - it just needs updating.
They do have a built-in monitor, and the iMac 5K is one of the best. At the price point for a prosumer computer + 5K display, it's a ridiculous good deal. I just got one -- what do I care if it has its own monitor? I'll use the thing for a bunch of years and then replace it, just like I would have with a stand-alone monitor anyway.
A bunch is... 3? 5? You replace your standalone monitors that often? What advances in tech do you need so often (what advances exist that often??), or what failures are you experiencing on the monitors, that you replace them so often?
JSG1 - Thanks for the cleanup. I'm no Engineer admittedly but love to catch up every now and then. A few days ago I was looking some PCI-Sig videos about what they're doing with version 4, 5 and 6. The roadmap has incredible bandwidth increases. I totally misspoke and alluded to PCI limitations with NVME. It's not the limitation of PCI that I really wanted to say was that a NVME SSD on PCIE4 doing 5Gbps is going to differ in real world results depending on usage case.
I for one and really glad Apple has a 6k Mac again. I'm old now...I remember when the IIFX was around $8k in a zero/zero config. For years Apple straddled the prosumer/workstation arena with PowerMac G4, G5 etc and many got used to these tweeners. The near future should be interesting as the Mac Pro internal design evolves with higher bandwidth solutions. It's never going to be cheap but if you make your living with that Mac you can often justify the purchase ...just hire a good accountant.
JSG1 let me ask you about your personal feelings about the future of tech like Optane.
No way to tell with Optane. Intel missed their target by a mile with the first release, and the next-gen Optane isn't a huge improvement. But... we've seen it before with Intel, where they keep iterating and eventually get to where they wanted to be. It's possible that Optane will yet live up to their promises. If so, then that's great, the world could really use a good SCM. If not, then it's a niche product, with some use for some people, but it won't move the whole market much.
SCMs in general though... They're a revolution waiting to happen. The big questions are:
1) Will they happen? Is there a tech coming that will really be able to replace both DRAM and flash? If I had to bet, I'd say yes, but I wouldn't feel very confident in that guess.
2) Will the price ever hit mainstream markets? No clue about that one at all.
Beyond SCM, there's in-memory compute. Not much like SCM, except for how it promises a major paradigm shift by changing how we think about memory. Interesting things happening there too, but it's too soon to try to predict where that's going. I think it's going to be big, though, unless it gets overtaken by something even bigger.
One thing I will say: The desktops we use today are not really different from the desktops we used ten or twenty years ago, seen from 10k feet. You have a CPU, memory, I/O busses, ports, video, storage. Details have changed a LOT but all the parts are organized in roughly the same way. I think that that will NOT be true ten years from now. Big changes are coming.
Why would you think that? I’d be really surprised if big changes actually come in ten (or even 20) years from now. The history you described is the real tell. The computer industry isn’t known to make big changes to hardware. Everything has been incremental. Any architectural change has to either be introduced in tiny steps or it isn’t introduced at all.
Tech people and programmers are historically known for being resistant to changing “how it’s done” in any way other than incrementally, and then there’s the bigger problem: late stage capitalism is afraid to invest in actually new things, especially if it’s expensive. There’s a fear of not being first with marketing and software, but, hardware with a fundamentally new design that isn’t based on existing cheap industry parts...?
It took forever to get high-PPI displays. That was an incremental (though long overdue) change driven by mostly one company pushing the rest of the industry with their “luxury” phones. Now that that company is back to being run by MBA-mentality-types, I don’t see much else changing any time soon.
I'm still pretty happy with my 2013 Trash Can. Might upgrade the SSD to 2 TB MacSales keeps lowering the prices. 32 GB ram is enough for me, I mainly use it for Capture One raw processing and Photoshop which it handles just fine. Been running 24/7 since new in 2013 with zero hiccups
I have the same machine for similar use, I'm now at 64 GB RAM (quite cheap now) and 1 TB SSD and two x 12 GB RAID 0 externally on TB2. That said I/O is way out of date and nothing we can do. My RAIDs 0 for Capture One 12 Pro and PS CC 2019 would be far faster with TB3. I have a lovely 27" 4K Dell monitor between two 27" Apple LCDs and the 4K is shared with my gaming PC when needed, so iMac is no use as no screen input change possible. My only path forward is a new Mac Pro. Given the 2013 Mac pro has lasted this long and hard to upgrade till recently, the 2019 Mac pro should last far longer than the 7 years I've waited. $6K seems a lot but if it lasts 10 years it's a steal.
Apple made its trash can Mac Pro limited by design. Even Apple could not upgrade anything in it! LOL! It is a total joke for both Apple and its owners! A good example of extreme proprietary design and function!
As I just mentioned above mine has been a fantastic Mac for the last 7 years. Zero regrets. I owned Cheese Graters before that and loved them too. It's only I/O making me upgrade now so yes lack of slots is now an issue but I want the latest Xeons and GPUs so all in all after 7 years I don't mind upgrading. Your argument about lack up upgradability is equally true of iMacs and Mac minis and Macbook Pros you know.
When I get the new Mac Pro I may turn the trash can into a good PC gaming machine, it runs Catalyst with Dual AMD Firepro 700s in tandem under Windows 10 Pro really well.
Why would you think that? I’d be really surprised if big changes actually come in ten (or even 20) years from now. The history you described is the real tell. The computer industry isn’t known to make big changes to hardware. Everything has been incremental. Any architectural change has to either be introduced in tiny steps or it isn’t introduced at all.
Tech people and programmers are historically known for being resistant to changing “how it’s done” in any way other than incrementally, and then there’s the bigger problem: late stage capitalism is afraid to invest in actually new things, especially if it’s expensive. There’s a fear of not being first with marketing and software, but, hardware with a fundamentally new design that isn’t based on existing cheap industry parts...?
It took forever to get high-PPI displays. That was an incremental (though long overdue) change driven by mostly one company pushing the rest of the industry with their “luxury” phones. Now that that company is back to being run by MBA-mentality-types, I don’t see much else changing any time soon.
The pessimist in me agrees with you. If we continue down a path of Computer Science grads taking their talents to FANG then we're not going to see much. We'll see hardware innovations but our software will continue to languish. I suspect someday I'll have a SSD or something like it that does 10Gbps and it'll still feel slow because the software is a decade behind the hardware.
We have to figure out a way to let the next generations go down different paths than the current. University-->Debt Slave-->would be Entrepreneur You're right though if any endeavor is burdened by the need for generate capital then clean slate approaches to computer science will not reach fruition.
I can't get over what a monster this thing is. Apple's "pro" machines have always been more marketed to prosumers/power users rather than true workstation users. This machine changes everything.
I'm not up on PC workstation class machines, so a question for someone who is: Is there anything even close to this?
Yes and no. You can get machines from HP, Dell, etc. with similar CPUs, RAM, and room for video cards. The afterburner card, no. That much Thunderbolt, no. Those video cards, no, though you can use NVidia cards that are either way better or somewhat worse, depending on what you do. Flash storage, yes, and you can do better than the Mac (FSVO better, again depending on use case). Slots... maybe not, I haven't checked.
Unfortunately, this Mac *still* hasn't shipped, whereas EPYC 2 is now readily available, with Zen-2 based Threadripper coming soon. While some people will still be unwilling to look at AMD products, I doubt that that will last long, as the AMD chips are ridiculously superior to Intel's product line, and will remain so for at least a year, I expect.
This Mac Pro will be a great workstation at a reasonably competitive price *for Intel-based workstations* (probably, but Intel pricing volatility and Apple's pricing stability may make Apple's pricing very unfavorable - time will tell). But cheaper Threadripper or EPYC-based workstations will wipe the floor with it, in most ways, and probably by the end of the year.
On a separate topic, does anyone know if the TB3 ports on the PCIe card are somehow provided with full bandwidth? Or are they constrained by the bandwidth of the PCIe card?
The I/O card slots into a half length x4 PCIe slot. x4 is enough to support 1 Titan Ridge TB3 chip, which supports the 2 TB3 ports on the card. And quite likely, the 2 TB3 ports at the top likely has their own Titan Ridge TB3 chip supported by x4 PCIe lanes. The iMac Pro, Mac mini and MBP w/TB are the same: 2 Titan Ridge TB3 chips for 2x2 = 4 TB3 ports.
If people upgrade to Apple’s higher end GPU options, they’ll get another 4 TB3 ports or more. The dual Pro Vega II GPU card supports 4 6K monitors. That has to mean their will 4 Titan Ridge TB3 controllers on the card. The machine will have more than it needs in terms of TB3 I/O if users buy either the Vega II or dual Vega II card.
As for AMD versus Intel, I think Apple values single thread performance a lot, and will tilt towards Intel as long as they have better single core performance. AMD will have more cores for cheaper, but like my other posts about a 2 CPU socket MPX module (or a 1 socket with a MCM package), Apple isn’t interested in winning a core count race. The up to 28 core Xeon W 3-series will be enough for the customers they are targeting.
The new Mac Pro will largely appeal to Audio and Video professionals who's company foots the bill. A ) it's not a Consumer machine and B ) it's under-featured and overpriced compared to PC Workstations from Dell, HP, and Lenovo...
A. Lots of YouTubers will buy them. These will be 2, 3, 4 person shops running FCP. You can maybe say there will be more sales to YouTubers than Hollywood studios. Another niche will be cloud service providers who will buy them for macOS hosting.
B. This point goes against what you are saying in A. These machines are in price tiers where the price of the hardware is a small part of the overall cost of computing. The big shops are paying hundreds of thousands of dollars in software licenses per year. They aren’t going to quibble of the cost hardware much, and it won’t be that different anyways.
JSG1 - Thanks for the cleanup. I'm no Engineer admittedly but love to catch up every now and then. A few days ago I was looking some PCI-Sig videos about what they're doing with version 4, 5 and 6. The roadmap has incredible bandwidth increases. I totally misspoke and alluded to PCI limitations with NVME. It's not the limitation of PCI that I really wanted to say was that a NVME SSD on PCIE4 doing 5Gbps is going to differ in real world results depending on usage case.
I for one and really glad Apple has a 6k Mac again. I'm old now...I remember when the IIFX was around $8k in a zero/zero config. For years Apple straddled the prosumer/workstation arena with PowerMac G4, G5 etc and many got used to these tweeners. The near future should be interesting as the Mac Pro internal design evolves with higher bandwidth solutions. It's never going to be cheap but if you make your living with that Mac you can often justify the purchase ...just hire a good accountant.
JSG1 let me ask you about your personal feelings about the future of tech like Optane.
No way to tell with Optane. Intel missed their target by a mile with the first release, and the next-gen Optane isn't a huge improvement. But... we've seen it before with Intel, where they keep iterating and eventually get to where they wanted to be. It's possible that Optane will yet live up to their promises. If so, then that's great, the world could really use a good SCM. If not, then it's a niche product, with some use for some people, but it won't move the whole market much.
SCMs in general though... They're a revolution waiting to happen. The big questions are:
1) Will they happen? Is there a tech coming that will really be able to replace both DRAM and flash? If I had to bet, I'd say yes, but I wouldn't feel very confident in that guess.
2) Will the price ever hit mainstream markets? No clue about that one at all.
Beyond SCM, there's in-memory compute. Not much like SCM, except for how it promises a major paradigm shift by changing how we think about memory. Interesting things happening there too, but it's too soon to try to predict where that's going. I think it's going to be big, though, unless it gets overtaken by something even bigger.
One thing I will say: The desktops we use today are not really different from the desktops we used ten or twenty years ago, seen from 10k feet. You have a CPU, memory, I/O busses, ports, video, storage. Details have changed a LOT but all the parts are organized in roughly the same way. I think that that will NOT be true ten years from now. Big changes are coming.
Why would you think that? I’d be really surprised if big changes actually come in ten (or even 20) years from now. The history you described is the real tell. The computer industry isn’t known to make big changes to hardware. Everything has been incremental. Any architectural change has to either be introduced in tiny steps or it isn’t introduced at all.
Tech people and programmers are historically known for being resistant to changing “how it’s done” in any way other than incrementally, and then there’s the bigger problem: late stage capitalism is afraid to invest in actually new things, especially if it’s expensive. There’s a fear of not being first with marketing and software, but, hardware with a fundamentally new design that isn’t based on existing cheap industry parts...?
It took forever to get high-PPI displays. That was an incremental (though long overdue) change driven by mostly one company pushing the rest of the industry with their “luxury” phones. Now that that company is back to being run by MBA-mentality-types, I don’t see much else changing any time soon.
I don't think that's exactly accurate. It is, over the short term, but the industry is more like the "punctuated equilibrium" school of darwinism: Things kind of roll along in an even fashion for a while, and then suddenly there's a phase change, and everything looks different in a couple of years. And then things stay the same again for a while.
Some examples: The GUI, and the mouse (and pointing devices in general). 32-bit (and then 64-bit) CPUs. LCDs displays. SSDs. Smartphones. DevOps.
I don't know that in-memory compute will be one of those things, but it has a good chance, because it may well get pulled into the mainstream as a side effect of being taken up by the HPC segment- it looks very promising for them. SCM is a certainty, *if* any of those technologies ever become cost-effective. Intel's pushing Optane hard, finding niches for it, and they may be able to push it all the way to mainstream if they keep at it. Spintronics are very promising. Memristors... who can tell, they certainly haven't made the splash I was hoping for. MRAM, PCM, Racetrack... there are a lot of possibilities in various stages of development.
Beyond this, I expect major changes due to the confluence of AR, VR, and AI. The biggest change I see coming is the end of the desktop paradigm, with windows, menus, etc. MacOS and Windows may or may not survive this change (both companies seem to be making strong moves, so if I had to bet, I'd say they will) - but if they do, you won't recognize them, as the UI will be totally different.
Walking further onto thin ice, I'll guess that the need for high-res high-refresh-rate heads-up displays will force a more distributed architecture, with lots of processing (mostly GPUish) in the display, some in input devices, and maybe NNPs decidated to voice and facial-expression recognition. But this is at best a SWAG.
Beyond that things get totally murky. I can imagine a future in which every room has lasers mounted in every ceiling corner, to overcome bandwidth limitations. Or possibly 100+GHz (or even THz) radio. In this scenario, there's no longer a cable from CPU to display. Intel and others have tried that in the past, with little success, but the needs of VR headsets may push us in that direction.
This reminds me of the Three Bears story! The Mac mini is too cold, the new MacPro is too hot! The 2013 is the not quite good enough!
Apple swung hard to the Animators & videographers with this design. Which is good!
But us photographers got short changed! Now if Apple where to take this design and make a desktop version dropping a few of the slots. Basically, the 2013 on steroids! I would buy one to replace my aging 2013 Trash Can!
You’re not the only one who wants a “less pro” Mac Pro, but there’s aren’t enough of you. Sure, Apple could make a cut-down version with fewer PCIe and DIMM slots, a smaller power supply and case, etc. But given the demand (relatively low) and buyer demographic (much less likely to order higher-end BTO configs), it would be priced somewhere around $9,999.
You're just asserting opinions as fact. You don't know what the demand is, and you don't know the "buyer demographic". That said, it's quite possible that Apple shares your opinions.
Your quoted price is ridiculous, though. Apple could do quite well with an "xMac" (as discussed here endlessly) in the $3k range. They just decided they'd rather segment their market in a different way.
Yeah, there is currently a $2500 to $5000 hole in Apple’s lineup right now.
The issue still seems to be Apple is offering the iMac 27” as its prosumer offering, and are foregoing any customers that want more. Don’t understand what they are thinking when they don’t offer up to 10 TB HDD options in the iMac 27, or 4 TB SSD options. They don’t need to change the design of the internals for the storage, just offer the option in the iMac 27. I’d get an 8 TB Fusion drive option in a heartbeat.
Then, if you want a 34” monitor, you are stuck with getting the Mac mini, and living with cabling hell. Heck, not selling a branded 27” 5K monitor to go with the Mac mini and TB3 laptops is definitely something I want to read about in a book. They basically left a billion dollars per year off the table, and I want to know what reasoning convinced them to not sell a TB3 5K monitor. And, it would have driven the iMac 27 down to a $1500 price tier.
They are still entirely too focused on limiting the product line, and many of the issues that got them in trouble in the 90s don’t really exist anymore.
This reminds me of the Three Bears story! The Mac mini is too cold, the new MacPro is too hot! The 2013 is the not quite good enough!
Apple swung hard to the Animators & videographers with this design. Which is good!
But us photographers got short changed! Now if Apple where to take this design and make a desktop version dropping a few of the slots. Basically, the 2013 on steroids! I would buy one to replace my aging 2013 Trash Can!
You’re not the only one who wants a “less pro” Mac Pro, but there’s aren’t enough of you. Sure, Apple could make a cut-down version with fewer PCIe and DIMM slots, a smaller power supply and case, etc. But given the demand (relatively low) and buyer demographic (much less likely to order higher-end BTO configs), it would be priced somewhere around $9,999.
You're just asserting opinions as fact. You don't know what the demand is, and you don't know the "buyer demographic". That said, it's quite possible that Apple shares your opinions.
Your quoted price is ridiculous, though. Apple could do quite well with an "xMac" (as discussed here endlessly) in the $3k range. They just decided they'd rather segment their market in a different way.
“Apple could do quite well with an "xMac" (as discussed here endlessly) in the $3k range.” Well I could say something snarky about you asserting opinion as fact, but that would be as ridiculous as when you said that to me. (Of course it’s my opinion, I’m the one who said it. I asserted nothing as fact.)
In any case, no Apple could not “do quite well with an xMac in the $3k range” because it would cannibalize sales to those who would otherwise buy a Mac Pro for $6k.
$6k works as the base price for the new Mac Pro only because the target market will typically order upgrades of one, more or all of the options: RAM, SSD, CPU or GPU. Who knows, but I’d expect the ASP of Mac Pro to be in the neighborhood of $12-15k—and those in the corporate/enterprise space will likely replace them every few years. (But the machine is excellent for small businesses or even one-person independent pros as well.)
PCIe and RAM sockets are cheap. Power supplies and cases are rather inexpensive as well. If the $6k Mac Pro has a BOM cost of $2k, a cut down version might be $1,700. And without Xeon and ECC, an “xMac” BOM cost might be, who knows, call it $1,000 (for all it matters).
Is it hard to see how a $6k entry level Mac Pro with a $12-15k ASP is a viable product, whereas a less expandable Mac Pro or a prosumer xMac (emphasis on the “sumer”) is a disaster? Not only would either potential offering save Apple very little—especially when you consider the full COGS, not just BOM cost—but the ASP would be crap.
Why do I say that? Simple logic. Anyone so desperate to save $50 per month (over 5 years) to spend $3k vs. $6k for the base machine is the same type of user who will buy their own RAM/SSD and even CPU upgrades. Apple will sell those customers one $3k base model—for a thousand dollars gross profit—every ten years.
80% of Mac customers buy laptops. That leaves 20% for mini, iMac, iMac Pro and Mac Pro. And I’d guess iMac is maybe 15 of that 20%. That leaves 5%, roughly a million units, split between the mini, iMac Pro and Mac Pro.
How many xMac can they sell? A few hundred thousand max. And given the effect of cannibalization of the Mac Pro (and to a lesser extent iMac Pro), it’s easy with a little modeling to see that Apple would lose revenue and profit by introducing either a less expandable Mac Pro or “prosumer xMac”. That’s true even if that new product were to result in a significant increase in units sold (which it likely wouldn’t since I think it would mostly steal demand from other models, rather than increasing the overall number of units demanded).
Anyone who needs a $3k base Mac Pro can surely afford a $6k Mac Pro. I don’t care what kind of pro you are: if $3k over a three or five or seven year lifespan is a dealbreaker, you’re doing it wrong. Close up shop and get a 9 to 5 instead. Those who want a less-pro Mac Pro can want one—that’s fine. But nobody needs one. Buy the $6k Mac Pro and sleep well at night knowing it has all the expansion capability you’ll ever need.
Stop trying to make xMac happen. It’s not going to happen.
Unfortunately, this Mac *still* hasn't shipped, whereas EPYC 2 is now readily available, with Zen-2 based Threadripper coming soon. While some people will still be unwilling to look at AMD products, I doubt that that will last long, as the AMD chips are ridiculously superior to Intel's product line, and will remain so for at least a year, I expect.
This Mac Pro will be a great workstation at a reasonably competitive price *for Intel-based workstations* (probably, but Intel pricing volatility and Apple's pricing stability may make Apple's pricing very unfavorable - time will tell). But cheaper Threadripper or EPYC-based workstations will wipe the floor with it, in most ways, and probably by the end of the year.
On a separate topic, does anyone know if the TB3 ports on the PCIe card are somehow provided with full bandwidth? Or are they constrained by the bandwidth of the PCIe card?
The I/O card slots into a half length x4 PCIe slot. x4 is enough to support 1 Titan Ridge TB3 chip, which supports the 2 TB3 ports on the card. And quite likely, the 2 TB3 ports at the top likely has their own Titan Ridge TB3 chip supported by x4 PCIe lanes. The iMac Pro, Mac mini and MBP w/TB are the same: 2 Titan Ridge TB3 chips for 2x2 = 4 TB3 ports.
That's what I think, too. That means you won't be able to push anything close to 40Gbps on both ports simultaneously. Rather, you'll have about 30gbps to share among both ports. Not surprising, but I had hoped for better.
If people upgrade to Apple’s higher end GPU options, they’ll get another 4 TB3 ports or more. The dual Pro Vega II GPU card supports 4 6K monitors. That has to mean their will 4 Titan Ridge TB3 controllers on the card. The machine will have more than it needs in terms of TB3 I/O if users buy either the Vega II or dual Vega II card.
4 ports, but I think only two controllers - I'm guessing that that's what's going to use the extra PCIe x8 connector in line with the x16 connector. Guess we'll see soon.
As for AMD versus Intel, I think Apple values single thread performance a lot, and will tilt towards Intel as long as they have better single core performance. AMD will have more cores for cheaper, but like my other posts about a 2 CPU socket MPX module (or a 1 socket with a MCM package), Apple isn’t interested in winning a core count race. The up to 28 core Xeon W 3-series will be enough for the customers they are targeting.
That may or may not be true, but you're behind the times when it comes to single-thread performance. AMD is no longer operating at a deficit.
Actually, when you get past 8 cores or so, single-threaded performance will always come second to core count, whether it's AMD or Intel. And I don't think your argument about "enough cores" is persuasive- if your workload scales to 28 cores, it will probably scale to 64 too. Beyond that, while price is not a top concern, AMD chips with equivalent performance and memory support to the Intel 28-core cost thousands less. That's significant.
In the short term I don't think this will matter much, but a year from now, I think it will matter more.
This reminds me of the Three Bears story! The Mac mini is too cold, the new MacPro is too hot! The 2013 is the not quite good enough!
Apple swung hard to the Animators & videographers with this design. Which is good!
But us photographers got short changed! Now if Apple where to take this design and make a desktop version dropping a few of the slots. Basically, the 2013 on steroids! I would buy one to replace my aging 2013 Trash Can!
You’re not the only one who wants a “less pro” Mac Pro, but there’s aren’t enough of you. Sure, Apple could make a cut-down version with fewer PCIe and DIMM slots, a smaller power supply and case, etc. But given the demand (relatively low) and buyer demographic (much less likely to order higher-end BTO configs), it would be priced somewhere around $9,999.
You're just asserting opinions as fact. You don't know what the demand is, and you don't know the "buyer demographic". That said, it's quite possible that Apple shares your opinions.
Your quoted price is ridiculous, though. Apple could do quite well with an "xMac" (as discussed here endlessly) in the $3k range. They just decided they'd rather segment their market in a different way.
“Apple could do quite well with an "xMac" (as discussed here endlessly) in the $3k range.” Well I could say something snarky about you asserting opinion as fact, but that would be as ridiculous as when you said that to me. (Of course it’s my opinion, I’m the one who said it. I asserted nothing as fact.)
In any case, no Apple could not “do quite well with an xMac in the $3k range” because it would cannibalize sales to those who would otherwise buy a Mac Pro for $6k.
$6k works as the base price for the new Mac Pro only because the target market will typically order upgrades of one, more or all of the options: RAM, SSD, CPU or GPU. Who knows, but I’d expect the ASP of Mac Pro to be in the neighborhood of $12-15k—and those in the corporate/enterprise space will likely replace them every few years. (But the machine is excellent for small businesses or even one-person independent pros as well.)
PCIe and RAM sockets are cheap. Power supplies and cases are rather inexpensive as well. If the $6k Mac Pro has a BOM cost of $2k, a cut down version might be $1,700. And without Xeon and ECC, an “xMac” BOM cost might be, who knows, call it $1,000 (for all it matters).
Is it hard to see how a $6k entry level Mac Pro with a $12-15k ASP is a viable product, whereas a less expandable Mac Pro or a prosumer xMac (emphasis on the “sumer”) is a disaster? Not only would either potential offering save Apple very little—especially when you consider the full COGS, not just BOM cost—but the ASP would be crap.
Why do I say that? Simple logic. Anyone so desperate to save $50 per month (over 5 years) to spend $3k vs. $6k for the base machine is the same type of user who will buy their own RAM/SSD and even CPU upgrades. Apple will sell those customers one $3k base model—for a thousand dollars gross profit—every ten years.
80% of Mac customers buy laptops. That leaves 20% for mini, iMac, iMac Pro and Mac Pro. And I’d guess iMac is maybe 15 of that 20%. That leaves 5%, roughly a million units, split between the mini, iMac Pro and Mac Pro.
How many xMac can they sell? A few hundred thousand max. And given the effect of cannibalization of the Mac Pro (and to a lesser extent iMac Pro), it’s easy with a little modeling to see that Apple would lose revenue and profit by introducing either a less expandable Mac Pro or “prosumer xMac”. That’s true even if that new product were to result in a significant increase in units sold (which it likely wouldn’t since I think it would mostly steal demand from other models, rather than increasing the overall number of units demanded).
Anyone who needs a $3k base Mac Pro can surely afford a $6k Mac Pro. I don’t care what kind of pro you are: if $3k over a three or five or seven year lifespan is a dealbreaker, you’re doing it wrong. Close up shop and get a 9 to 5 instead. Those who want a less-pro Mac Pro can want one—that’s fine. But nobody needs one. Buy the $6k Mac Pro and sleep well at night knowing it has all the expansion capability you’ll ever need.
Stop trying to make xMac happen. It’s not going to happen.
I agree with a lot of your analysis, and more to the point, it looks like Apple may also.
About opinions vs. facts - the opinion I was calling out was that the demographic is so small. What I offered as fact ("could do quite well") wasn't clear enough, sorry - I meant, they could make a reasonable profit per unit. (We know this because other brand-name sellers do it just fine.) I wasn't saying that it would fit in with their current segmentation, and I wasn't saying it wouldn't cannibalize other models.
Despite your well-argued reasoning, I still think Apple is making a mistake. They're leaving some sales on the table, and I think you could design the product so it wouldn't cannibalize much of the high-end market - carefully chosen limits on RAM and PCIe expansion should leave buyers of the xMac happy while not being adequate for the typical buyer of a $12-15k nnMP.
They would certainly cannibalize some sales of iMacs and iMac Pros. That's a good thing. The world doesn't need that many LCD panels out there, from an ecological standpoint, and Apple's incremental margin on the iMac's screens is not that large.
Guys I wouldn't be too sure that Apple isn't going to deliver xMac or something that fits that Prosumer hole @Tht speaks of. I feel today's hardware has gotten us to a point but going forward there are some things that aren't sustainable. I just had to add $50 to my Cox plan because all TV in my home are now 4K and I was hitting my bandwidth cap by week 3. Broadband providers understand they have the power because if I'm going to continue streaming I need to go unlimited. Now I can continue to pay the extra fee or I can begin to look at larger storage options.
Widescreen displays are growing more and more popular. 27" just feels really confining today. How do we address this without entry level limitations on one end of the spectrum and overbuying on the other end?
Apple's revenue and profits don't come from the Mac so these aren't "bet the company" decisions. If I'm Apple I've long since realized that a key of disinter-mediating the movements of broadband providers must come with hardware solution linked to Apple Services.
I'm not convinced that Apple is out of the router game. I think the router is simply going to become a more multifunctional device. I think ARM based Macs have the opportunity to come in non-standard form factors with easy setup
I'm totally for a modern Mac with a bit of internal expandability roughly the size of this that doesn't demolish the savings account.
If people upgrade to Apple’s higher end GPU options, they’ll get another 4 TB3 ports or more. The dual Pro Vega II GPU card supports 4 6K monitors. That has to mean their will 4 Titan Ridge TB3 controllers on the card. The machine will have more than it needs in terms of TB3 I/O if users buy either the Vega II or dual Vega II card.
4 ports, but I think only two controllers - I'm guessing that that's what's going to use the extra PCIe x8 connector in line with the x16 connector. Guess we'll see soon.
I’m thinking the Titan Ridge chip only has enough bandwidth for 1 5K or 6K monitor. Hence why the MBP15 only supports 2 5K monitors. Not sure if the MBP15 will have enough GPU perf to drive 2 6K monitors, but I don’t think it won’t be the Titan Ridge controller holding it back.
The Pro Vega II card in the Mac Pro supports 2 6K monitors, which implies 2 Titan Ridge controllers on the card if it is 1 Titan Ridge chip per 5K or 6K monitor. The dual Pro Vega II card supports 4 6K monitors. Really can’t see how 2 Titan Ridge controllers can support that many.
As for AMD versus Intel, I think Apple values single thread performance a lot, and will tilt towards Intel as long as they have better single core performance. AMD will have more cores for cheaper, but like my other posts about a 2 CPU socket MPX module (or a 1 socket with a MCM package), Apple isn’t interested in winning a core count race. The up to 28 core Xeon W 3-series will be enough for the customers they are targeting.
That may or may not be true, but you're behind the times when it comes to single-thread performance. AMD is no longer operating at a deficit.
Actually, when you get past 8 cores or so, single-threaded performance will always come second to core count, whether it's AMD or Intel. And I don't think your argument about "enough cores" is persuasive- if your workload scales to 28 cores, it will probably scale to 64 too. Beyond that, while price is not a top concern, AMD chips with equivalent performance and memory support to the Intel 28-core cost thousands less. That's significant.
In the short term I don't think this will matter much, but a year from now, I think it will matter more.
It looks to me that Intel still has the single threaded performance crown. It’s smaller now. For the Xeon W 3000 series and the Mac Pro and Threadripper, we will have to wait and see how much power AMD is willing to burn in the Zen 2 Threadripper models.
Don’t get me wrong, I’d like Apple to go with a Threadripper for the Mac Pro, but it’s not going to be as much of win for AMD as you think. Intel will and has started to price their CPUs to get into price/perf parity with AMD as AMD takes more and more market share. The Xeon W 3000 series will be perfectly fine for the Mac Pro. Over the grand scheme of things, the best multicore performance doesn’t look that important to Apple. They just want enough for the Mac Pro, and they put all their bets in the GPU MPX modules and the Afterburner cards.
Also, I didn’t say 28 cores was enough. That’s what Duhsesame thinking I did. I was the one wondering if Apple could make a MPX module with 2 CPU sockets, or 1 gigantic socket for MCM CPU packages that both Intel and AMD use.
It's worth noting the "custom SSD" is the same module from the iMac Pro. It's not really an SSD, just a raw flash interface with no controller. The T2 lives on the logic board. Processor to T2 is PCIe x4, T2 to the flash chips is this specialized connector.
In theory, this would make aftermarket SSDs easier to manufacture, because you don't need to source a controller. You just need to wire the board properly and have a bunch of raw flash chips. In practice, I doubt we will see aftermarket SSDs. Nobody else is using this type of system, so there's no reason for other companies to make bulk storage for it.
I wonder how the T2 will deal with capacity changes. I assume it will just wipe both flash carts.
Nah, right now the SSD is locked by the T2, having slots doesn't make any difference.
Unless Apple does unlock the firmware, but even so you may not be able to swap the controller.
Except the flash carts aren't soldered in place, so they are clearly replaceable. So the question is how the system handles the replacements. As Apple has full control over the controller, they could create a system to rearrange how the actual flash data is used.
Most of the hard work is already done. I've dealt with servers which support hot-swap RAM and hot-add RAM. You tell the OS you want to remove a particular RAM book. It evacuates the contents of that book, then tells you it is safe to pull. You pull the book, replace a bad stick (or increase the capacity, either by filling empty slots or replacing existing sticks), reinsert the book, tell the OS it's okay to use again, and the OS starts sending data back to it.
Fundamentally, flash storage isn't any different. Tell the drive controller to evacuate certain chips, notify the user when it's done, and there you go.
This wouldn't have come up with the iMac Pro, since even though it has two flash carts, it's not meant to be disassembled by the user, so they don't have to deal with flash cart swaps.
Um, you "paused" the OS, then swap an SSD with different files.....
I think it will cause system-wide destruction.
That said, being physically possible (well, you can also desolder the NAND too) doesn't mean you CAN do it, certainly not the iMac Pro. I hope they do, but so far they don't.
You're not actually reading anything I wrote, are you? Goodness.
There are two flash carts. TWO.
Apple could write software to let the user tell the T2 they want to replace one of the two flash carts.
The T2 could then move all the data off of the flash cart you want to replace, leaving you with one flash cart and an empty slot.
You could then add a second cart to the empty slot. The software could even potentially detect a larger flash cart and give you more space.
All of this could even be done without shutting down the OS, though I bet Apple forcibly shuts down the machine when you open the case.
Wow just twigged , this as an old repost, my comments on this thread are way out of date and prior to a rethink.
I don't see that many posts from you on this thread. The other one in which you/Madan/Mel exchanged plenty of posts, that is the good one for nostalgia i guess.
Wow just twigged , this as an old repost, my comments on this thread are way out of date and prior to a rethink.
I don't see that many posts from you on this thread. The other one in which you/Madan/Mel exchanged plenty of posts, that is the good one for nostalgia i guess.
This page 4 has two from me that I immediately saw that made me realize. I did not go back any further. Yes, I eat humble pie regarding some exchanges. My new iMac 5K 27" i9 with upgraded GPU and my own added 64 GB RAM rocks. I simply don't need a workstation now I am retired.
I will add that since then and reading the current Mac pro 2019 post IMHO the nomenclature is wrong and Apple could have avoided all the prosumer angst and envy by naming this Mac Workstation rather than Mac Pro ... just my 2 cents.
This reminds me of the Three Bears story! The Mac mini is too cold, the new MacPro is too hot! The 2013 is the not quite good enough!
Apple swung hard to the Animators & videographers with this design. Which is good!
But us photographers got short changed! Now if Apple where to take this design and make a desktop version dropping a few of the slots. Basically, the 2013 on steroids! I would buy one to replace my aging 2013 Trash Can!
You’re not the only one who wants a “less pro” Mac Pro, but there’s aren’t enough of you. Sure, Apple could make a cut-down version with fewer PCIe and DIMM slots, a smaller power supply and case, etc. But given the demand (relatively low) and buyer demographic (much less likely to order higher-end BTO configs), it would be priced somewhere around $9,999.
You're just asserting opinions as fact. You don't know what the demand is, and you don't know the "buyer demographic". That said, it's quite possible that Apple shares your opinions.
Your quoted price is ridiculous, though. Apple could do quite well with an "xMac" (as discussed here endlessly) in the $3k range. They just decided they'd rather segment their market in a different way.
That's an opinion, and there's no basis for it in reality, but there is the fact that Apple has chosen not to make the elusive xMac I've been hearing about for decades. It's funny how I hear people claim about greedy Apple is while at the same time also hearing people say how Apple is leaving so much on the table. Outside of everyone's opinions as to whether Apple should or shouldn't make an xMac or how wildly popular and successful it would be for Apple, the fact remains that they don't. At some point you just have to stop your bellyaching. Now, if we're done with wishing for impossible things, can I get back to writing a letter to Fox to un-cancel Firefly?
Comments
Why would you think that? I’d be really surprised if big changes actually come in ten (or even 20) years from now. The history you described is the real tell. The computer industry isn’t known to make big changes to hardware. Everything has been incremental. Any architectural change has to either be introduced in tiny steps or it isn’t introduced at all.
Tech people and programmers are historically known for being resistant to changing “how it’s done” in any way other than incrementally, and then there’s the bigger problem: late stage capitalism is afraid to invest in actually new things, especially if it’s expensive. There’s a fear of not being first with marketing and software, but, hardware with a fundamentally new design that isn’t based on existing cheap industry parts...?
It took forever to get high-PPI displays. That was an incremental (though long overdue) change driven by mostly one company pushing the rest of the industry with their “luxury” phones. Now that that company is back to being run by MBA-mentality-types, I don’t see much else changing any time soon.
When I get the new Mac Pro I may turn the trash can into a good PC gaming machine, it runs Catalyst with Dual AMD Firepro 700s in tandem under Windows 10 Pro really well.
You're right though if any endeavor is burdened by the need for generate capital then clean slate approaches to computer science will not reach fruition.
As for AMD versus Intel, I think Apple values single thread performance a lot, and will tilt towards Intel as long as they have better single core performance. AMD will have more cores for cheaper, but like my other posts about a 2 CPU socket MPX module (or a 1 socket with a MCM package), Apple isn’t interested in winning a core count race. The up to 28 core Xeon W 3-series will be enough for the customers they are targeting.
A. Lots of YouTubers will buy them. These will be 2, 3, 4 person shops running FCP. You can maybe say there will be more sales to YouTubers than Hollywood studios. Another niche will be cloud service providers who will buy them for macOS hosting.
B. This point goes against what you are saying in A. These machines are in price tiers where the price of the hardware is a small part of the overall cost of computing. The big shops are paying hundreds of thousands of dollars in software licenses per year. They aren’t going to quibble of the cost hardware much, and it won’t be that different anyways.
The issue still seems to be Apple is offering the iMac 27” as its prosumer offering, and are foregoing any customers that want more. Don’t understand what they are thinking when they don’t offer up to 10 TB HDD options in the iMac 27, or 4 TB SSD options. They don’t need to change the design of the internals for the storage, just offer the option in the iMac 27. I’d get an 8 TB Fusion drive option in a heartbeat.
Then, if you want a 34” monitor, you are stuck with getting the Mac mini, and living with cabling hell. Heck, not selling a branded 27” 5K monitor to go with the Mac mini and TB3 laptops is definitely something I want to read about in a book. They basically left a billion dollars per year off the table, and I want to know what reasoning convinced them to not sell a TB3 5K monitor. And, it would have driven the iMac 27 down to a $1500 price tier.
They are still entirely too focused on limiting the product line, and many of the issues that got them in trouble in the 90s don’t really exist anymore.
In any case, no Apple could not “do quite well with an xMac in the $3k range” because it would cannibalize sales to those who would otherwise buy a Mac Pro for $6k.
PCIe and RAM sockets are cheap. Power supplies and cases are rather inexpensive as well. If the $6k Mac Pro has a BOM cost of $2k, a cut down version might be $1,700. And without Xeon and ECC, an “xMac” BOM cost might be, who knows, call it $1,000 (for all it matters).
Is it hard to see how a $6k entry level Mac Pro with a $12-15k ASP is a viable product, whereas a less expandable Mac Pro or a prosumer xMac (emphasis on the “sumer”) is a disaster? Not only would either potential offering save Apple very little—especially when you consider the full COGS, not just BOM cost—but the ASP would be crap.
How many xMac can they sell? A few hundred thousand max. And given the effect of cannibalization of the Mac Pro (and to a lesser extent iMac Pro), it’s easy with a little modeling to see that Apple would lose revenue and profit by introducing either a less expandable Mac Pro or “prosumer xMac”. That’s true even if that new product were to result in a significant increase in units sold (which it likely wouldn’t since I think it would mostly steal demand from other models, rather than increasing the overall number of units demanded).
Widescreen displays are growing more and more popular. 27" just feels really confining today. How do we address this without entry level limitations on one end of the spectrum and overbuying on the other end?
Apple's revenue and profits don't come from the Mac so these aren't "bet the company" decisions. If I'm Apple I've long since realized that a key of disinter-mediating the movements of broadband providers must come with hardware solution linked to Apple Services.
I'm not convinced that Apple is out of the router game. I think the router is simply going to become a more multifunctional device. I think ARM based Macs have the opportunity to come in non-standard form factors with easy setup
I'm totally for a modern Mac with a bit of internal expandability roughly the size of this that doesn't demolish the savings account.
It looks to me that Intel still has the single threaded performance crown. It’s smaller now. For the Xeon W 3000 series and the Mac Pro and Threadripper, we will have to wait and see how much power AMD is willing to burn in the Zen 2 Threadripper models.
Don’t get me wrong, I’d like Apple to go with a Threadripper for the Mac Pro, but it’s not going to be as much of win for AMD as you think. Intel will and has started to price their CPUs to get into price/perf parity with AMD as AMD takes more and more market share. The Xeon W 3000 series will be perfectly fine for the Mac Pro. Over the grand scheme of things, the best multicore performance doesn’t look that important to Apple. They just want enough for the Mac Pro, and they put all their bets in the GPU MPX modules and the Afterburner cards.
Also, I didn’t say 28 cores was enough. That’s what Duhsesame thinking I did. I was the one wondering if Apple could make a MPX module with 2 CPU sockets, or 1 gigantic socket for MCM CPU packages that both Intel and AMD use.
There are two flash carts. TWO.
Apple could write software to let the user tell the T2 they want to replace one of the two flash carts.
The T2 could then move all the data off of the flash cart you want to replace, leaving you with one flash cart and an empty slot.
You could then add a second cart to the empty slot. The software could even potentially detect a larger flash cart and give you more space.
All of this could even be done without shutting down the OS, though I bet Apple forcibly shuts down the machine when you open the case.
I will add that since then and reading the current Mac pro 2019 post IMHO the nomenclature is wrong and Apple could have avoided all the prosumer angst and envy by naming this Mac Workstation rather than Mac Pro ... just my 2 cents.