It comes with a tiny price that can be overcome by memory upgrades down the line. We’re in the early days of DDR5 and LPDDR5 still, a few years down the line my M1 Max will have exactly the same RAM it does now, but my PC laptop could be running 4x the ram with significantly faster speed and lower latency. Guess which one will likely have a faster memory system after the couple hundred dollar upgrade?
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 May Not Launch Updated Mac Studio With M2 Ultra Chip Due to Similarity With Upcoming Mac Pro
- Thread starter MacRumors
- Start date
- Sort by reaction score
Only needs a slight upgrade to overcome the tiny loss having DIMM slots rather than soldered memory causes. Plus, it’s really common for PC motherboards/CPUs to support much much much faster ram with a one click “over clock” (XMP) than they officially support. My current PC tower motherboard came out in early 2019 when ~3200mhz CL16 RAM was considered fast, but can easily “officially“ support memory at least as fast as 4200mhz CL16 now. That’s at least 30% faster with a similar drop in latency. In bandwidth terms, 2 sticks of DDR4 3200 have about 50GB/s theoretical bandwidth, while 2 sticks of 4200 have a theoretical max bandwith of almost 70GB/s. If you can show that soldering RAM to the package gains over 40% over the same socketed RAM, I’ll eat my hat live on video.
Apple not only integrates the SoC in a single die, but it has the memory chips on the same package. This allows them to use low power mobile DDR chips, and since they are on package it also reduces significantly all the power that having the memory transactions run through the system's PCB externally would consume.
Where do you think Apple will get the 22 hour battery life from?
When people make comparisons between say a 2023 MBP 16" M1 Max 5nm 140W vs any 2023 Intel Core i9-13900hx 10nm 330W based laptop they all forget to compare the battery life, power consumption, thermals and other metrics that contributed to that raw performance.
All devices being brought onto aircraft are cannot go beyond 100 watt hour.
Apple's current logic board design vs all other desktop/laptop design allows Apple to do away with most of the capacitors on their package, by putting them straight onto the other side of their die. This makes for a better PDN (their cores use, ironically a lot of power but for short bursts) and reduces system cost.
This is the front and back of the MacBook Pro 16" (2021 A2485) 10-Core CPU/16-Core GPU Logic Board with Paired Touch ID Sensor
Pls point out where you would put any of those expansion slots that Apple could easily add to?
I became acquitted with this logic board recently as I was wondering why Apple could not put a M1 Ultra SoC into a MBP 16" as it only consumes another 100W of power when the newest USB PD charger standard of 240W was released May 2022.
Many pointed out to me that maybe the M1 Ultra SoC is just too big and too hot.
I am having difficulty finding the LPDDR5-6400 SODIMM. Could you help me find any?
Last edited:
I wouldn't call 3200 to 4200 with better timings more than a slight upgrade and I think benchmarks/real world would agree.
Maybe if you're pushing through huge tasks daily and those few seconds really count. Which I guess the Mac pro would be focused on but still seems like not worth mentioning to me.
Biggest advantage of user replacabe RAM/SSD is users not paying Apple's absurd premiums and ability to change later.
Apple is well aware of that and has very little incentive to allow it. It's the opposite clearly.
Different logic board design. There are many PC laptops with smaller volume than the 16” MBP that have space for SO-DIMM slots and NVMe.Apple not only integrates the SoC in a single die, but it has the memory chips on the same package. This allows them to use low power mobile DDR chips, and since they are on package it also reduces significantly all the power that having the memory transactions run through the system's PCB externally would consume.
Where do you think Apple will get the 22 hour battery life from?
When people make comparisons between say a 2023 MBP 16" M1 Max 5nm 140W vs any 2023 Intel Core i9-13900hx 10nm 330W based laptop they all forget to compare the battery life, power consumption, thermals and other metrics that contributed to that raw performance.
Apple's current logicboard design vs all other desktop/laptop design allows Apple to do away with most of the capacitors on their package, by putting them straight onto the other side of their die. This makes for a better PDN (their cores use, ironically a lot of power but for short bursts) and reduces system cost.
This is the front and back of the MacBook Pro 16" (2021 A2485) 10-Core CPU/16-Core GPU Logic Board with Paired Touch ID Sensor
Pls point out where you would put any of those expansion slots that Apple could easily add to?
I became acquitted with this logic board recently as I was wondering why Apple could not put a M1 Ultra SoC into a MBP 16" as it only consumes another 100W of power when the newest USB PD charger standard of 240W was released May 2022.
M1 Ultra SoC is just too big.
I am having difficulty finding the LPDDR5-6400 SODIMM. Could you help me find any?
As for the power savings of on-package LPDDR, they make sense for like a MBA. And can even be argued as ”worth it” on the MBP though I tend to disagree. But for even the Mac Mini, let alone the Studio and a theoretical Pro? Saving a few watts to vastly limit the long term potential of the machine is a really bad idea. There’s no reason Apple’s SoCs can’t run more traditional RAM, except that Apple won’t let them. This is a benefit in smaller devices, but a major negative in the larger form factors. It all goes back to the fact that apple is the world’s best mobile computing device company, but they have no interest or ability to play outside of that sandbox.
We all understand that Apple has a huge financial interest in seeing how just how ****** they can make their machines for consumers before we stop buying. That doesn’t change the fact that you’re dismissing a 40% bandwidth increase (larger than the M1 to M2 jump) as a worthless upgrade. It’s far from it for many applications. If your use case doesn‘t see a benefit from a major RAM bandwidth increase, you had more computer than you needed to start (Nothing wrong with that, but there are plenty of users out there who are always looking for more bandwidth or other “small“ gains that pay for themselves many times over in their work)
All those PC laptops have inferior
- die shrink
- transistor count
- raw performance
- performance per watt
- thermals/waste heat
- battery life
- power consumption
- GPU & CPU core counts
And no equivalent other features to Apple Silicon
So you want Apple to redesign a whole series of Apple Silicon chips just to accommodate your design targets?
x86/Apple Silicon desktops make up ~20% of all PCs shipped annually worldwide.
Worldwide PC shipments totaled 286.2 million units in 2022.
Of which 27.911 million units are Macs.
~20% of that is ~5.5822 million Mac desktops that are mostly Mac mini M1.
Economies of scale dude... economies of scale. I did not bother computing how many Mac Studio much less Mac Pro get sold.
That is why I question the commercial viability of a Mac Studio with PCIe expansion slots at a starting price of $6k based from the 2019 year model.
Mac Studio users probably represents >50% of Mac Pro users who have zero requirement for PCIe expansion slots. Why else spend any R&D money to develop such a "gimped" Mac Pro if there is not demand for it? Apple made a gimped Mac Pro in 2013-2017, 2017-2020 & 2022-today.
A traditional Mac Pro with PCIe expansion slots was missing from 2013-2019.
Lower units shipped = increased per unit cost.
>80% of Apple Silicon's R&D money came from iPhone chips. That's why the M series are SoC that shares the same design targets of a smartphone. They do not have upgradeable RAM and SSD because of targets of
- raw performance
- battery life
- power consumption
- lower material manufacturing parts
- space constraints
Apple is a system's vendor. Meaning that they sell the finished product, not just the processors. So they can use several parts from the vertical process to subsidize others. In this case, Apple can afford to make very good SoCs because they don't sell those chips elsewhere, meaning that they are not as pressured to make them "cheap" in terms of area for example. Since they're going to recoup the profit from elsewhere in the product.
In contrast; AMD and Intel sell their processors to OEMs, so they only get profit from the processor not the finished system. So they have to prioritize cost, by optimizing their designs for Area first and then focus on power. This is why both AMD and Intel use smaller cores, which allows them for smaller dies. But which have to be clocked faster in order to compete in performance, unfortunately that also increases power.
This is probably their key difference; Apple can afford the larger design that is more power efficient for the same performance. Whereas AMD/Intel have to aim for the smaller design that is less power efficient for the same performance.
Your use case is very very valid but may not be at at the worldwide annual shipment volume that makes it profitable or economical for Apple to pursue further product refresh without major changes in material cost.
It is like the 2023 HomePod. Its predecessor was pulled from the market after a price cut as it wasn't working out. It returned when they reengineered it for better material cost.
Last edited:
As in the front end website. Is it owned by Microsoft?
I've seen volume corporate Office licensed mis-licensed to retail channel.
It is a legit copy but not intended for retail buyers like you or me.
These are things Apple can fix by investing in their chip design.
The concerns (and wild speculations) come from it being pretty obvious from others that there isn't a good economical reason for Apple to make that investment. They can use their current SoC tech for the basis of every chip from a bluetooth headphone to a pro laptop. The only device that they'd support with that (IMHO massive) investment is the Mac Pro.
That said, there are technologies that do make sense that would make their SoC tech more robust and potentially more economical, such as how AMD is doing chiplets on mixed processes. I suspect eGPU support over thunderbolt 4 is coming (if not already there in the M2 Pro/Max), and a chiplet design may also help in creating alternate designs with significant PCIe lanes. Thats not to imply I think migrating their current interconnect to support chiplets and mixed hardware budget (e.g. 2/2 CPU/GPU or 1/4 CPU/GPU population in the chip depending on the hardware needs)
But, we don't get much exposure to Apple Silicon rumors - there's no third party parts suppliers leaking information, there's nobody taking money on the assembly line. So we don't _know_ much of anything.
What we can see is the M2 Max hasn't added in the modularity to match the current flexibility of the Mac Pro.
Pls post links to these specific CPU benchmarks.
Also link relevant Idle & CPU Max figures in terms of
- Watts
- BTU/hr
How many % of 2019 Mac Pro users uses all those expandabilty options?
R&D of the 2013 Mac Pro, 2017 iMac Pro & 2022 Mac Studio may tell a different story.
I am NOT claiming your stated use cases are invalid. I am pointing out that it may not be at an annual units shipped worldwide quantity to encourage Apple towards annual refresh like those from 2006-2013 Mac Pro.
At what volume does AMD sell Threadripper Pro chips? What are the projected number of Mac Pro users with similar/same hardware requirements?
Are you a customer for that AMD product? If yes then how many others are out there?
There are reasons why Apple did not prioritize the gaming PC industry. And it has to do with leveraging their tech and seeing if it is profitable for them to put in any money into it.
Intel hamstring them from 2006-2020. With Apple Silicon they are a bit more confident.
The design targets you point out are very boutique.
I would be surprised if more than a 100 thousand out of 286.2 million units of PCs shipped worldwide in 2022 were the predecessors of those tech specs.
What I have observed with user replies are based on their past experience with Intel/PPC Macs and Windows PCs.
Apple is applying how the flagship smartphone makers does business onto the Mac and their other product lines.
The reason why Apple has the efficiency edge is mainly due to Apple having an edge in terms of the node they use, the PDN (power delivery network) tech, and packaging.
Their ARM cores are actually more complex than the x86 competitors; significantly wider and with larger resources for out of order and speculation. Most people assume there is some kind of "magic" that makes ARM better that x86, but that is not the case. The ISA has little impact on overall power consumption given the same microarchitectural resources.
Apple uses their larger/more complex cores to their advantage, by running them at a slower clock rate. While allowing them to do more work per clock cycle. This allows them to operate on the frequency/power sweet spot for their process. One has to note that power consumption increases significantly (way higher than linear) the higher the frequency.
Here is where the PDN technology comes into play. Apple uses the most advanced technology to distribute power to keep all the functional units feed, which requires the ability to supply a lot of instantaneous power. To do so, Apple uses a 3D stacked architecture of 2 dies; one for the logic, and another one on top (or bottom depending where you look at it) to distribute the power. In contrast, almost every one else has to use the same die to do logic and distribute power.
The irony is that a simpler/smaller ARM core would have to be clocked faster in order to compete with Intel/AMD cores. And it would end up consuming the same high power.
Apple also has a very good SoC design. Meaning that they integrate most of the system on a single die; the CPUs, the GPU, the NPU (AI accelerator), the Codec (video processing), the camera block, I/O (USB, WiFi, ethernet, PCIe/TB, etc), and the memory controller.
For some stuff like AI and video encoding, having custom silicon handling it is far far more efficient than running it on a general purpose code.
Lastly, it also comes to packaging. Apple not only integrates the SoC in a single die, but it has the memory chips on the same package. This allows them to use low power mobile DDR chips, and since they are on package it also reduces significantly all the power that having the memory transactions run through the system's PCB externally would consume.
So it's a combination of Apple using a single package where Intel/AMD laptops require multiple through their PCBs to support the same functionality. As well as Apple having access to better overall fabrication technology for that single package that AMD/Intel have for theirs.
The trend seems to be that it is becoming more efficient for mobile vendors to scale up their products into laptops, than it is for desktop vendors to scale down their products into laptops.
There is also a key difference in business models: Apple is a system's vendor. Meaning that they sell the finished product, not just the processors. So they can use several parts from the vertical process to subsidize others. In this case, Apple can afford to make very good SoCs because they don't sell those chips elsewhere, meaning that they are not as pressured to make them "cheap" in terms of area for example. Since they're going to recoup the profit from elsewhere in the product.
In contrast; AMD and Intel sell their processors to OEMs, so they only get profit from the processor not the finished system. So they have to prioritize cost, by optimizing their designs for Area first and then focus on power. This is why both AMD and Intel use smaller cores, which allows them for smaller dies. But which have to be clocked faster in order to compete in performance, unfortunately that also increases power.
This is probably their key difference; Apple can afford the larger design that is more power efficient for the same performance. Whereas AMD/Intel have to aim for the smaller design that is less power efficient for the same performance.
Apple needs to have a stark differentiator between their products and those based on Intel/AMD or else there is no justification for the price premium.The interesting thing is that the single thread for any future Apple Silicon system is going to be 90%+ available on Apple’s low end Air. Imagine that, for any single threaded processes that aren’t long running (>10 minutes), users of Apple’s lowest end systems will be within a stone’s throw of the high end.
And, rather than being amazed at that reality, folks will still wonder why Apple doesn’t intentionally create less performant mobile processor solutions (like the competition does) so that the performance difference appears bigger.Mobiles just aren’t supposed to provide any performance metrics close to the high end! That’s what Intel/AMD’s been telling us for years, it MUST be right… right?
A lot of the increase of efficiencies is because of replicating smartphone design methods onto a laptop and desktop.
Example of which is to put the SSD controller onto Apple Silicon SoC. It reduces material manufacturing cost, latency and power overhead. They then put the NAND memory chips very close by on the logic board so they can easily BTO that.
What also keeps Apple Silicon SoC cheaper is that the cost of R&D is mostly (>90%) paid for by iPhone sales. Whatever Mac-specific features is then financed by sales of the Mac (<10%).
They have this arrangement because Mac gross profit cannot finance Mac chip R&D at the level of those of Intel/AMD. It would be more economical to use iPhone chip-derived tech and add Mac-specific tech requirement.
Apple is a mobile device company. That’s what I’m saying. They need to stop pretending they cater to a Pro or high performance market. If they’re not willing or able to do the work to build a competive high end system, be ok being the best mobile device maker in the world. Lots of people find the MBP or even just an iPhone meets all their computing needs. The isssue I have With apple is that they pretend they care about a high performance market when they clearly haven’t for over a decade. The 2019 Mac Pro was a flash of potential, and the Apple silicon transition could have been great, but it’s clear (and understandable!) that they want to focus on the mobile market exclusively. Fine, stop pretending and just do that. I’ll move my high performance systems to another OS and keep a Mac for iMessage and web browsing with maybe some light photo/video work on the go.
Owned and registered by Microsoft; yes.
Individual not volume corporate or other.
Legit as in registered and downloaded via Microsoft official website.
You seem skeptical as if you have been burnt in the past, I am interested in hearing your story relating to said experience.
Link pls? I've seen those ads of cheap Office and Adobe perpetual licenses on Facebook ads. The image looked like it was produced by a student on MS Paint.
Apple identified that >50% Mac Pro users do not need PCIe express slots and just want the fastest Mac for the money.
Hence
- 2013 Mac Pro
- 2017 iMac Pro
- 2022 Mac Studio
Your interpretation may be better stated that extreme use cases are slowly dwindling to a point that it isn't economical to R&D such products further at a cadence in the past.
It is simply a demand issue and not Apple they " don't care". Their care is inversely proportional to the quantitative units sold annually.
It is like how mainframe computers dominated before Y2K. Fast forward quarter of a century later and who still sells computers the size of a football field?
So they reworked workstations/servers into supercomputer clusters to replicate or closely clone those boutique requirement.
You may want to consider the drop of PC shipment in 2022 as compared to 2020 & 2021. Odds are the next 4-6 years it will further drop as the bulk of forced upgrades was co-terminus with COVID and are returing back to a cadence of pre-2020.
As nodes improve the demand for workstation desktops decreases. Not saying there isn't at least 1 person requiring it for a very unicorn use case but it may be more economically served by using a more general purpose desktop or even laptop.
The tech has steadily improved hence a ~80% laptop & ~20% desktop PC ratio. That was the main motivator of Steve Jobs to move from PPC to Intel as he saw years prior to 2005 that that will be the future of the Mac.
There is a market for a Mac Pro with PCIe express slots but it was being subsidized by Mac users who want the most powerful Mac for the money.
With the cheaper & more powerful Mac Studio without PCIe express slots the number of buyers for a Mac Pro is cut by more than half.
As fewer units are expected to be sold will force Apple to agressively negotiate with parts supplier to lower cost or increase the base MSRP from $6k to more than that or a combination of those two. The severity of which would be based on how many units Apple expects to sell them.
In theory Apple could disable the memory controllers of the RAM and memory chips and use a more traditional DIMM or M.2 slot if they wanted to.
But due the the limited volume that is to be expected it may go beyond $6k to accomodate that because of redundancy brought in by having to use the 3rd party's built-in memory controllers that comes with the SSD & DIMM.
All I can say is if Apple solely used the Mac Pro profits to fund these modifications it will cost more than $6k base.
Last edited:
Apple has nearly 100bn per year in profits. AMD’s entire market cap is 130bn, and intel’s is under 120bn. Apple can afford to develop more chips than both AMD and intel combined and barely even put the line item on their statements. Apple is freaking huge. They just don‘t care to spend any money on anything other than mobile devices and services. That’s fair. Developing dedicated GPU dies, 40+ core CPU dies, and entire new platforms like 64 or 128 bit wide LPDDR memory modules is not worth it when they can ream consumers with obscene upcharges at the point of purchase and have an army of lapdogs running around claiming that the “innovation” of screwing over consumers for profit somehow makes their overgrown mobile phone SoCs better for high end computing than much higher performing hardware from other companies. Everyone reasonable agrees that Apple builds the best mobile devices in the world for most users. But they don‘t serve the high performance computing market. No amount of spin or wizardry can make any version of M2 Super ultra Max compete with true high performance hardware. And that’s ok. All I’m asking is that they either build a Mac Pro or retire the name. Releasing a “Mac Pro” with anything related to the rest of the Apple Silicon lineup is just an insult.
Servers do more computing than ever before, what are you smoking? Most of AMD, intel and Nvidia’s profits come from datacenter and supercomputer sales, not consumer hardware. Apple has been trying and failing to force the idea of a “pro level” machine with no expandability down their customers throats for over a decade, and all they’ve succeeded in is driving most of their bigger accounts away. Nothing wrong with that, they make a lot more money from iPhones than they ever could from Mac Pros, and business is business. I just wish they’d stop pissing on Pros and calling it rain.
GPU GPU GPU. How many times do I have to say it. Because I can (with an extra power supply) run 2x Radeon 6800XT in a 2009 Mac pro, it can run circles around the overgrown cell phone GPU in the Studio Ultra. Geekbench open CL for just one 6800XT is 105k in MacOS, while the Studio Ultra manages about 75k. A 2019 Pro can have 4 of those GPUS, for literally 5 times the compute performance. Plus, what if you work with data sets more than about 100GB? That’s swap city on even a maxed out Studio, or using up like 8% of a Maxed out 2019 Pro’s RAM.
Also efficiency doesn’t really matter for high end desktop. The number of people who need more than laptop performance but can’t afford a couple bucks a year ($.2/kWh * 250 watts extra on average * 6 hours/day * 250 workdays per year = $75 per year worst case. Before we even calculate lost time waiting for the Mac or lost income due to slower completion of work. If you’re using the power professionally, it pays for itself almost instantly.)
AMD threadripper pro is closely related to AMD EPYC at this point. And those chips sell as fast as they can be built (hence why TR pro only comes out every other generation and zen 4 Threadripper is coming out nearly a year after zen 4 for desktop, laptop and server). We all agree that HEDT is not a large part of the market, but AMD and intel sell server chips so they can toss a few dies to the HEDT users when needed. Apple only sells cell phone chips that can’t scale past “decent desktop performance with great efficiency.”
Apple needs to abandon all pretense of being a company that makes “pro” hardware or they need to develop server grade chips and go after that market while also releasing workstation/HEDT products.
Finally, it’s just embarrassing to pretend that no one used PCIe slots just because you don‘t. Even with the 2019 Mac Pro being only 3 years old, there are so many uses for those slots, and as time goes on there will be more and more. Off the top of my head, if I had a 2019 Mac Pro I’d run a PCIe to 4x NVMe card, a FireWire card (still use that when messing with old MiniDV tapes), probably 2x4 slot GPUs, and that only leaves like one slot for the future (probably would be used for either a future Thunderbolt revision or 40+ Gb Ethernet when I build up my network storage solution a bit more. By the time the 2019 Mac Pro is truly end of life, with its second or third user, probably sometime in the 2030s, who knows what cards will be relevant? One thing I do know, by 2032, a 2022 Mac Studio will be museum pieces or curiosity compared to what any 2019 tower will have been upgraded to handle.
Apple did not pursue the server market further. I was hoping they'd make a ARM cloud to increase volume of Apple Silicon chip like the Ultra (two Max chips) or even the rumored Extreme (four Max chips).
Many here make snide remarks about performance per watt and power consumption. When it comes to data centers, server farms and supercomputers the $/kWh and monthly electricity bills impacts the bottom line.
That's why three high-ranking engineers from Apple's chip division formed Nuvia to pursue without Apple.
They then got bought out by Qualcomm for them to work on the Android/Windows equivalent of Mac chips.
A bit ironic that they came back to where they started.
Similarly to any person's home. Most people again wave aside such efficiencies but if all your lighting, appliances, computers, devices and anything else that uses the grid were all as in all of them reduced power consumption by nearing 70% then your monthly kWh use would drop by that amount and the cost of that specific line item would cut by nearly 70% as well.
So when I say I want Apple silicon efficiencies I also look at replacing any incandescent bulb, floricent lamp and LEDs pre-5 years ago with lighting fixtures with the efficiencies of 223 lumens per watt as that materially lowers the utility bills.
In my mind all said electrical equipment should have the lowest operating thermal output possible so the HVAC would only cool the persons inside it assuming insulation keeps the heat out the room. Going any lower would require the people in said room to be dying or dead so we do not omit any BTU/hr.
My switching from a 2 decade old twin tub top loading washer and ringer to a front loading inverter washing machine reduced my water consumption by half and the power bill improved as less water was used that also related to less water weight to moved around inside the washing machine. It also helped reduce the need for detergent as well.
Someone made a snide remark about PCMaster race types with 1.5kW PSUs with nearing 80% efficiency. They only get away with that because a lot of them (not most) still live with their parents who pay them power bills.
If they paid their own bills, had a spouse and kid(s) to feed, insurance to pay, morgage, retirement then they'd not be putting and RGB LED lighting anywhere. lol
Apple saw the workflow of their Mac Pro customers from 20012-2013 and that guided them to create
- 2013 Mac Pro
- 2017 iMac Pro
- 2022 Mac Studio
It's a quantitative approach.
If Mac Studio gets a M2 Max/Ultra before the M3 is released in 2024 then it is an indicator that
- a lot of people bought the Mac Studio
- there's a surplus of M2 Max chips from the MBP 14"/16" and Apple needs to put it somewhere
Same reason iPhone mini did not get an iPhone 14 refresh.
Demand may be there but not in a quantity that merits an annual product refresh.
Hence 2019 Mac Pro remained unchanged for 3+ years.
iPhone SE refreshed in these years
- 2016
- 2020
- 2022
- 2024 (?)
Last edited:
You are commenting on unreleased products that may probably outperform or at least match the currently released hardware at a lower TDP.
How many Mac Pros with PCIe express slots would have been sold if the Mac Studio was
- there
- not there
What you described could be replied to with these questions
Quantify the number of users or units sold that have that workflow.
How many non-synthetic apps can take advantage of said hardware setup.
Apple has these in terms of sales figures and when macOS reports back to Apple whenever it queries about a compatible macOS Software Update.
I ask as Nvidia's SLI tech was notoriously difficult to code for in relation to number of users who actually used it.
Regarding AMD's business practices
- https://wccftech.com/amd-admits-to-...last-two-quarters-plans-to-continue-practice/
- https://www.gizchina.com/2023/02/03/amd-plans-to-overprice-cpus-gpus-despite-underselling-them/
- https://www.fudzilla.com/news/pc-hardware/56269-amd-admits-undershipping-to-keep-prices-high
Links above may relate as to how many users are there for Threadripper Pro.
I never claimed no added/modified their Mac Pros available PCIe expansion slots. I call into question if there are >50% Mac users who did. As the development of the 2013 Mac Pro, 2017 iMac Pro and 2022 Mac Studio indicates to me <50% demand for it.
You have a very unicorn workflow. I do not blame you for being this disturbed by Apple catering to your exacting standards.
I also have 3 miniDV video cameras in cold storage. Should have sold it before 2008. Just too lazy to transfer the footage to my Mac and liquidated them when it had some worth.
Personally? I'd love to have vacant M.2 slots + the soldered NAND memory on the logic board in all the Macs that actually have space for it.
Though I would not trade off the performance improvements of having RAM on the SoC package.
Totally agree with these 2 statements.
There is no way you can get Nvidia's levels of performance on the GPU if it is integrated and shares memory with the CPU. It has to be dedicated and suck up huge amounts of power and produce heat.