Which would be better served by going with the Mac Studio right?Media work, photo and video.
Which would be better served by going with the Mac Studio right?Media work, photo and video.
Depends on what you're doing and what software you're using. For most people working at home on small and mid size projects, sure. Others might need x86 compatibility (Rosetta 2 can't emulate all x86/Intel instructions), dual boot into Windows or need more RAM than 128GB. You're also missing out on the installation of PCIe cards for additional storage, I/O and so on, if needed. No dual 10GBit ethernet and the lack of ECC memory on the Studio (this is a killer feature). Also, the NVMe drives in non MacBook Macs are bound to cause trouble right now: https://threadreaderapp.com/thread/1494213855387734019.html (another killer)Which would be better served by going with the Mac Studio right?
Is it enough of a performance boost over what was offered before for Apple to bother writing drivers for compatibility, knowing the current Mac Pro is going to be the last/only Mac to support the hardware?Depends on what you're doing and what software you're using. For most people working at home on small and mid size projects, sure. Others might need x86 compatibility (Rosetta 2 can't emulate all x86/Intel instructions), dual boot into Windows or need more RAM than 128GB. You're also missing out on the installation of PCIe cards for additional storage, I/O and so on, if needed. No dual 10GBit ethernet and the lack of ECC memory on the Studio (this is a killer feature). Also, the NVMe drives in non MacBook Macs are bound to cause trouble right now: https://threadreaderapp.com/thread/1494213855387734019.html (another killer)
Or it could simply be for compute work, but that could probably be done on some cloud server running Nvidia GPUs.
The Mac Studio is not a Mac Pro replacement, it's an alternative for low end Mac Pro configurations. It caps out at around $8k vs. $50k+ for a fully loaded Mac Pro. Those are two completely different machines, different use cases. Apple has yet to introduce their full Mac Pro replacement.
I’m not sure the source for your numbers so I decided to do a bit of digging:Snip
With the W6800 Duo and W6900? Sure, some people need the performance. As far as drivers, that's a question for AMD, not Apple who support them.Is it enough of a performance boost over what was offered before for Apple to bother writing drivers for compatibility, knowing the current Mac Pro is going to be the last/only Mac to support the hardware?
Wait, AMD writes macOS drivers? I thought Apple was handling it in house. I wonder why they don't do monthly updates to the drivers like they do in Linux/Windows (mainly Windows, lol).With the W6800 Duo and W6900? Sure, some people need the performance. As far as drivers, that's a question for AMD, not Apple who support them.
AMD does "base drivers" and Apple integrates into macOS, they're working together. You can't download a driver directly from AMD, it comes with macOS, hence the lack of frequent updates. I think you can download bootcamp drivers from AMD directly, but it's been ages since I've last used bootcamp, so not sure.Wait, AMD writes macOS drivers? I thought Apple was handling it in house. I wonder why they don't do monthly updates to the drivers like they do in Linux/Windows (mainly Windows, lol).
I wonder why Apple doesn't allow the vendor to do more frequent updates.AMD does "base drivers" and Apple integrates into macOS, they're working together. You can't download a driver directly from AMD, it comes with macOS, hence the lack of frequent updates. I think you can download bootcamp drivers from AMD directly, but it's been ages since I've last used bootcamp, so not sure.
Quality control... if there is such a thing with Apple these days.I wonder why Apple doesn't allow the vendor to do more frequent updates.
What's even the purpose for all the thunderbolt ports of the Mac Studio if the intent is for it to be a completely closed appliance? Are they there essentially for cosmetic purposes?I don’t understand why they bothered releasing the W6000 series at all on the Mac pro.
External storage and the thunderbolt monitor they just released, please pay attention.What's even the purpose for all the thunderbolt ports of the Mac Studio if the intent is for it to be a completely closed appliance? Are they there essentially for cosmetic purposes?
They haven’t deprecated support for the 7,1?It does seem fairly foolish to have built the Mac Pro then W6000 module if they were going to do an about face to that path by going completely counter standards depreciating users machines a few short months/years later
They’ve assigned people to contribute to the Blender project and routinely show off Maya in their keynotes.and then staying as stubborn as possible by thumbing their nose at the Mac 3D community and prospective 3D Mac users.
They’re popular in audio work, especially with room for AVID cards.The Mac Pros were predictably going to sell poorly given the price point and lack of Nvidia hardware still and anyone working in any area of media knew they were extreme overkill for video and photo work.
See above.It seems like it was all a pointless sham rather than a reorienting and broadening of use case scenarios for Macs for other neglected areas like 3D.
So the people making the Metal frameworks are idiots?Apple forming a strategic Alliance with Sony/Playstation would be ideal as I doubt Apple has any high level 3D tool or content developers on the MacOS team.
Or that in making Optix, NVidia has formed a moat around the segment.Tim Cook also doesn't really seem technically inclined so he's not an advocate for any of it, if you work in 3D in any area everyday, you're always left wondering whether the reason Apple hasn't made significant headway with 3D for any category, relative to 3D on Windows, is because they just don't have the people for it.
So you suggest making an entirely new API? That’s beyond asinine. Moreover, AMD gpus in the Playstation share very little hardware similarities with the Apple Silicon gpus. There’s no expertise to be found.If a strategic alliance with Sony was created Apple could have Mark Cerny contribute to Mac development and help craft a more inclusive set of graphics APIs that facilitate more content that is already made and dependent on other industry standard hardware at this point to getting to the Mac and the hardware running on the Mac via thunderbolt based eGPUs.
Ah yes, Google glass, that huge wearable success. Has Google had that much success in hardware? I know I haven’t seen anything that convinces me they have a leg up on Apple.This move alone would re-engage third party software that's been abandoned actually ported to the Mac quickly while still maintaining MacOS' future goals outside of 3D and building towards complete Metal native operation in the future. Apple could also hire Richard Marks from Google and he could contribute to hardware development in a number of areas including Apple wearable tech in general.
Microsoft begs to differ.For the people that claim their only concern for "AAA" games being on the Mac are that they are risky and costly: Playstation is the number one brand in gaming.
Give a single good reason Sony would do this.A strategic alliance with Sony/Playstation would increase opputunities for more hardware sales and mitigate any potential risk that might be incurred since Playstatio finances all it's own projects.
Lol “minimal” you clearly have no knowledge of these costs.If there were any losses (which wouldn't happen) the alliance could be terminated, parties go their separate ways and minimal money would be lost.
See above.Apple could also hire Shawn Layden, Shannon Studstill and Alan Becker as "old hands" in the gaming business that could push 3D content and OS development at Apple past the low hanging fruit Apple Arcade level and that would be informative for games development as well as general development enhancements for other areas in 3D for Apple.
Just to be clear I never questioned the money coming from the mobile sector. ALL my numbers came from statista. I question the sustainability of the market given the extremely low number of contributors to it relative to the whole market (only 10%.)I’m not sure the source ...
That's 2 devices not 5 and frankly as simple as those they could easily be daisy chained on 1 thunderbolt port not 5.External storage and the thunderbolt monitor they just released, please pay attention.
That’s a possibility, but I’d still bet on that remaining the lion’s share of the market, and it remaining a bigger source of revenue.Just to be clear I never questioned the money coming from the mobile sector. ALL my numbers came from statista. I question the sustainability of the market given the extremely low number of contributors to it relative to the whole market (only 10%.)
More specifically I question the long term viability of that revenue stream considering Apple won't be getting it's customary 30% take from that minuscule group as litigation is forcing Apple to open secondary purchase sales to outside groups they can't collect all that money from.
Apparently you haven’t been paying attention to the desktop game market, which adopted those strategies a long time ago.Above all of that I question the ethical nature of the mobile games sales in general as the games that do actually make money are designed to target and psychologically manipulate people with obsessive/compulsive tendencies and disposable income, so called "Whales" to spend, spend, spend.
If this ever comes to fruition it’ll be a shakeup on the entire market. And the entire gaming market is ethically bankrupt. See: sexual harassment lawsuits, regular employee burnout, and other fun things like day 1 paywalled content which were a big deal back in the day but gamers were just conditioned to accept!Apple has always purported itself to be an ethical company, both directly and indirectly. The nature of the mobile games business is going to be further scrutinized publicly at some point as to whether it should be regulated under the rules of gambling proper and gambling has always existed in a grey area ethically.
Am I reading this correctly and you’re taking Epic’s side?If Apple were to be drawn into any legal battles involving the mobile games business and gambling directly it could likely hurt the company's reputation. Considering how ruthless Apple had been about not wanting to foster competition of any kind in the Epic vs. Apple litigation in the mobile games space that would be viewed as a green to all manner of psychological manipulation for the purposes of generating profit.
The studio is not a Mac Pro replacement.Unless a person needs to run Windows (and not even then really) the value to performance to price proposition with what the Mac can run well isn't there to even consider buying a Mac Pro right now, with the Mac Studio available.
The RED/ARRI community begs to differ, as the link above to the forum post also suggests.The full capacity of the Mac Pro exceeded it's usefulness in audio and video production using the applications it was best suited to run for most well equipped studios. The Mac Pro has pretty much been the Bizarro of creative workstations.
Uhm, no. Maybe start here: https://www.raywenderlich.com/books/metal-by-tutorials and then compare to other APIs and hardware. You can always add translation layers though, but at the cost of performance. Also Apple isn't interested in this and the hardware doesn't support it.Nope, never stated that. Metal, the MacOS and Apple Silicon can be expanded/extended to fully interoperate with Nvidia and/or AMD GPUs at any time.
The studio is not a Mac Pro replacement.
1.) Memory is limited to 128GB, which is nothing compared to the Mac Pro. It is more than enough for what most people do at home and the average YouTuber making videos. It is by far not enough for professional users in the industry, working on $100M+ projects.
2.) No ECC memory in the Studio, which is a deal killer. Search for that old story when Apple sold non-ECC professional machines doing compute jobs for weeks and hundreds of machines kept crashing and had to be replaced later with ECC machines. No ECC in the enterprise world is a no go.
I go back to Magma PCIE to PCIE enclosures and I was actually ecstatic with their early adoption of thunderbolt. Thunderbolt is very robust as a pro solution and a modernization update to firewire. It's really Intel's handiwork of artificially impeding Thunderbolt's adoption after Apple turned it over to them that has kept it from being widely adopted as a replacement in a lot of different data transfer cases. SDI and XLR are stone aged jumbo sized "pro" ports for instance that should have been phased for locking thunderbolt ports and adapters years ago, but that's a different discussion for a different time.4.) Studio can't be easily upgraded. Thunderbolt ports are for "home users" and do their jobs, but how can one add additional NVMe controllers, or low latency I/O, HD-SDI, etc.?
These are optimization issues where outside of the codecs Apple is (hopefully) at the end of a blood feud with RED and Blackmagic over RAW codec patents (A recurring motif for Apple). As a result RED and Blackmagic have been slow to optimize their RAW decoders on M1. They're going to have to do it and soon though. There are too many of their customers on Macs that will switch to Assimilate/Final Cut/ProRes as a full end to end replacement for RedRAW, BRaw etc. at a drop of a dime where that lagging performance doesn't exist.5.) Raw power, as usual Apple took specific benchmarks to let the Studio shine, while for other things is much slower than a Mac Pro. You don't have to go that far into Hollywood dubstages/studios, right here in this forum is enough: https://forums.macrumors.com/threads/mkbhd-switching-to-mac-studio-from-mac-pro-but-why.2340919/
So a W6800x Duo equipped Mac Pro is 60% faster than a fully maxed out Studio Ultra when working with R3D Raw. If one is making money with such a machine, I'd sure know what I'd buy right now. And I'd also know what I'd buy as a hobbyist doing some work at home, even if it makes me a little money.
Uhm, no. Maybe start here: https://www.raywenderlich.com/books/metal-by-tutorials and then compare to other APIs and hardware. You can always add translation layers though, but at the cost of performance. Also Apple isn't interested in this and the hardware doesn't support it.
"...the hardware doesn't support it."
Case by case in the sense that when the enduser reading emails at home and browsing the web won't need more than 128GB RAM? Sure. Same for the hobbyist cutting home videos and touching up photos from grandmas last birthday party.Case by case. (this is going to be a recurring motif)
No, it it's. You just don't understand the technology, that's all. Normally I would ask what "Apple's Virtual RAM scheme" is, to reduce it and would expect you to explain the algorithm. But since so far you have failed to answer a single technical question and reply with marketing nonsense instead, it's pointless because by now everyone knows you don't have the technical background to answer it. You probably don't need it and that's ok, it's just that normally when someone has no clue about it, they choose not to argue about things they don't understand. In psychology it's called selective response.Sure, there are are circumstances where more memory is preferable, but the combination of SSD optimization and Apple's Virtual RAM scheme reduces the "Absolutely gotta have it" cases by a lot.
No one said anything about servers. The Mac Pro is what it is, a professional workstation, similar to what Dell, HP and Lenovo offer.In larger operations it's just not going to be a server room "Run by Mac Pros" scenario.
No they don't. Check the modules used and compare datasheets of off-the-shelf chips you can order in bulk. Also, higher quality (by what metric, things can be measured you know) has nothing to do with ECC.Another case by case situation. On the content creation side where Macs are concerned (and as long as you avoid Gen 1 of anything Apple) Apple tends to source a dramatically higher quality memory and drives to begin with.
Enterprise data storage and retrieval have nothing to do with the need for ECC. Again, you don't understand the technology of ECC, which again is taught to first semester computer science students. Normally people would try to understand what ECC actually does.For Enterprise where there's the high levels of data storage and retrieval and where they have to be on Windows or Linux or both currently I can see where they must use ECC.
What an utter nonsense. Where do you come up with these google results. It's quite clear you've never been in broadcast and movie industry where this is the standard. And yes, the hobbyist and YouTuber at home doesn't care about it. Professionals in $$$ projects do.SDI and XLR are stone aged jumbo sized "pro" ports for instance that should have been phased for locking thunderbolt ports and adapters years ago, but that's a different discussion for a different time.
And this has what to do with what I said? Nothing, right. A NAS in any form doesn't replace local storage, leave alone I/O for interfaces. As for PCIe enclosures... you're taking a massive performance hit and since you specifically mention Magma as a manufacturer, check what they do to PCIe lanes (datasheets and measurements, you know...) ... funny you choose this Manufacturer of all for the professional market when it's all about performance.Faster CPUs/GPUs and fast NAS clusters that got easier to maintain along with with highly optimized OSes running on SSDs made those Magma enclosure obsolete, but thunderbolt has continued to progress well and is far beyond anything most home customers need for external peripherals.
No, they're not. I've worked at the research institute that came up with h264 around the same time when Amir Majidimehr worked on VC-1 at Microsoft. Again, normally I'd ask for a technical explanation (and not marketing material by fanboys/sales stuff) why you think that is the case. But since you're not able to answer it, I'll just leave this. There is always room for optimization, but not on this large performance gap. M-series is not the magic some people make it out to be. It's all about performance-per-watt. ProRes is so much faster than on x86 machines because M-series chips have specific hardware for ProRes, they lack this specific hardware for other codecs. The same is true for photo work, there's a nice article and thread over at dpreview (for the end user). That doesn't mean Apple won't introduce optimized hardware with M2+ in the future, but for M1/Pro/Max/Ultra that ship is sailed.These are optimization issues where outside of the codecs Apple is (hopefully) at the end of a blood feud with RED and Blackmagic over RAW codec patents (A recurring motif for Apple).
Indeed...LOL...
When all you do is making claims and not answer a single technical question and choose selective responses instead to avoid delivering answers, that's what you get. You don't even understand what the difference between a driver and a full blown toolkit is from your google search. That alone unfortunately shows you don't have any understanding about the underlying software or hardware. What else would one recommend other than read up on the very basics of technology.Ok, I'm always open to digging in and learning new things as a lot of people are, but the mere fact that everybody has to toss out a lot of years of experience and knowledge gained just to get to baseline, low level functionality on the Mac is offensive.
So you're willing to take a performance hit for GPU applications that makes it slower than a native CPU implementation? I'm not, because you know... it's even slower than without a GPU.The performance hits of translation layers are preferable to a draconian system that eliminates choice.
As you said above, LOL indeed. Is that all you could come up with in your google search?More importantly Otoy, the company behind the Octane renderer has been talking about CUDA code translation for a number of years. Apple brought many of their engineers into Apple for some publicly unspecified development before the M1 was released. I never could suss out what they were doing then, but now looking on it it wouldn't surprise me if Apple had Otoy build them a super efficient CUDA translation layer for Mx SOC processors. That would explain a lot of things actually. We'll have to wait and see.
"According to Otoy’s CEO, Jules Urbach, the point of developing this CUDA translation layer is so that the company’s high-end Octane Render software can run as easily on AMD GPUs as their Intel counterparts. “We have been able to do this without changing a line of CUDA code, and it runs on AMD chips,” Urbach said. “You can now program once and take CUDA everywhere. AMD has never really been able to provide an alternative.”
So yes, LOL indeed to what you posted. LOL indeed.The latest release of Octane X is the capstone of years of development to rebuild the industry’s leading unbiased GPU path tracer from the ground up, optimized for maximum performance with the Apple Metal API. Octane X features full pixel parity with OctaneRender 2020 and later, making state of the art cinematic rendering available natively for the first time in macOS.
This sentence alone shows you don't understand the technology. Maybe do a search what actually happens to the CUDA code when using translation.WE WILL HAVE ACCESS TO THE CUDA CODES...bwa..hahahahahaha
Case by case in the sense that when the enduser..
Yeah, since I'm not responding in the way you would prefer it's probably indicative of those descriptions not being useful in practical applications discussion.it's just that normally...
I know what ECC, how the coding works, what what it's uses are and that it originated in radio communication. Frankly that's more than adequate as it relates to practical use of the hardware with pipeline specific software and specific groupings or hardware to practically determine if it's a worthwhile purchase option for your use and for your groups uses.is taught to first semester computer science students....
Sounds like you've never run cable or transported heavy cable knowing there are modern lighter weight digital cables that can be easily re-purposed for most uses as they are in other areas and asked the question, "Why are we still using these dinosaur analog cables from the 1980's and earlier?" (and that could have been from the early 2000's all the way to today) It's very easy to not ask questions when you just take orders, see the things and leave those things for other people to contend with.What an utter nonsense....
Don't worry, I work with industry partners all the time. You on the other hand do not, not on a professional level as you have demonstrated so often now by posting false stuff. Nice try though.One area of difference between an old University Comp Sci. Professor without day to day practical hands on creative industry experience and someone that actually works in the creative industry (and has for years) is that the person with the industry experience in a number of different capacities with multiple degrees knows that coders are told to shoot for the stars while the person working in the trenches, seeing how the sausage is made realizes that 80% of the various specialty areas that contribute to various final "$$$" products don't need as lofty a spec as what the coders are told to build to get their day to day tasks complete.
I know you have yet to post/answer anything technical that is remotely correct. That's good enough for this forum, you know nothing and are wrong all the time as demonstrated. That driver/toolkit thing was on a new level, but a really good laugh.I guess you can keep projecting your personal anecdotal descriptions onto me like you know anything about me since it seems help you out in some way, weird as that is.
You're not responding because you don't know what you're talking about and have no answers, because you don't even understand the questions to begin with, just like you don't understand the difference between a driver and a toolkit as demonstrated. As I said, it's called selective response in psychology and it speaks volumes.Yeah, since I'm not responding in the way you would prefer it's probably indicative of those descriptions not being useful in practical applications discussion.
As you have evidently demonstrated, no you don't.I know what ECC, how the coding works, what what it's uses are and that it originated in radio communication.
Because SDI is so analogue and Thunderbolt cable length allow for high distances... huh? Another thing you just demonstrated the lack of knowledge. It's also odd none of the major broadcast facilities, studios and dub stages want this. But hey, nice try again trying to go off topic.Sounds like you've never run cable or transported heavy cable knowing there are modern lighter weight digital cables that can be easily re-purposed for most uses as they are in other areas and asked the question, "Why are we still using these dinosaur analog cables from the 1980's and earlier?"
Don't bother. It's quite clear you need time to come up with new google results you don't understand without answering any questions to your claims (which you're actually required to by forum rules) only to go off topic further to draw people away from the false stuff you posted so far. As you have demonstrated multiple times now that you have no clue about this stuff, the conclusion is simple... you're wrong and you're not getting anything you want, no matter how much you whine about it. @JMacHack already nailed you spot on why you're posting this.I'll go through the rest of this at some point later in the week
Mac is getting Resident Evil Village. Seems like Apple talked to Capcom (maybe threw some money their way?).
Wonder if Capcom will do Monster Hunter in MacOS.
Yeah that game didn’t do well on the other platforms so this feels like it is destined to do poopily.Capcom is trying to test the market with REVIII. If it does well on MacOS, then it is a possibility.