New M4 Pro Mac Mini vs M2 Ultra Mac Pro vs PC Threadripper - We Test Them!

[@samsudbrink9085]

This has to be the best M4 breakdown on KZbin. Thank you

[@DavesGarage]

Wow, thanks!

[@ralf391]

@@DavesGarage this review makes all those gasping-in-awe Apple sycophants look like idiots. But hey, the kids have to make money somehow. A mention of power draw for all machines and the price of the AMD behemoth would have been nice.

[@PKperformanceEU]

No its not. This was hands down the most x86 biased comparison of the week! No 7995wx is 10x faster than m4pro or m2ultra 🤡 3x faster in reality! Just wait for m4extreme It will crush even a OC 2000watt 7995wx Arm64 v9 is a superior platform to slow garbage x86

[@gravelrhoads]

@@PKperformanceEU I'm guessing you're just a troll.

[@paulstubbs7678]

@@PKperformanceEU Only if your into multicore, in which case you know what you need

[@TomGreen99]

Just don't go speed testing a Lamborghini in a school zone.

[@Stalkingwolf]

I get this reference

[@deavo74]

@ same lol.

[@DavesGarage]

Just FWIW, the Tesla video was filmed on a closed drag strip. Car hit 161 in the quarter!

[@deavo74]

@ I feel the reference was aimed at a certain channel with around 19.6m subscribers who since edited out the said reference. You are safe lol. 161 by the way!?!?

[@deavo74]

Also why has KZbin got so odd when replying to people and remove their name.

[@RbNetEngr]

Great review, Dave. I appreciate your clarity for presenting this information, and the short but sufficient explanations to help viewers understand the significant takeaways from each set of results.

[@DavesGarage]

Glad it was helpful!

[@reyne5746]

Dave! So refreshing to listen to someone speaking sense about benchmarks and making a common sense comparisons. Fantastic.

[@alxxer]

Finally a proper benchmark, I've been looking for.

[@DaveWhiting01]

Let's not forget that the Threadripper is a $5K CPU with a power draw of 500 Watts

[@Prosper85]

its the power draw thats stupid , Apple are covering all bases with their designs.

[@akshat8586]

​@@Prosper85They're slow.

[@JusticeItself1]

*10,000

[@CharlieMartorelli]

Excellent comparison, best on youtube. Just the facts no bias. Thanks for the work you put in to this.

[@DavesGarage]

I appreciate that!

[@ray_glaze]

I am reminded of the old joke... All machines wait at the same speed.

[@DavesGarage]

I've never heard that before, but I'm stealing it!

[@ReadersOfTheApocalypse]

How fast is your car in a traffic jam?

[@Cyril29a]

@@DavesGarage Picasso said great artists steal because he understood that stealing was making something one's own. It isn't about depriving others of the original but rather modifying the original so much that it becomes a new thing that now belongs to the thief.

[@SaHaRaSquad]

@@Cyril29a True except Picasso didn't say that, it's just what everyone thinks after Steve Jobs said that once. The original quote existed before Picasso was born.

[@PKperformanceEU]

@@DavesGarageonly clowns ever use r23 on arm64 devices!!!!!!🤡🤬

[@NickLaslett]

Dave, did not expect this kind of video from you. But this is easily the most entertaining and informative video I’ve seen covering the new Mac mini. Well done.

[@pembrokejones3969]

As someone who paid over $125000 for an Avid editing system in the 90's, I am amazed what you can get now for $1500! my first Avid in 1988 used 600 megabyte r-mag drives that cost $2500 each and you needed 2, as well as a chassis that also cost $2500!

[@roberthuff3122]

🎯 Key points for quick navigation: 00:00 *🖥️ Introduction and Competition Overview* - Introduction of devices being compared: M4 Pro Mac Mini, M2 Ultra Mac Pro, and a high-end PC. - The purpose of the video is to test these devices in performance benchmarks. 01:15 *🚗 Personal Tech and Comparison Motivation* - Personal anecdote about cars leading into the discussion on technological superiority at home. - Comparison motivation between the M2 Ultra Mac Pro owned by the host and his wife's new M4 Pro Mac Mini. 01:58 *💰 Mac Mini Cost Efficiency* - Discussion on the cost-efficiency of the base model M4 Mac Mini. - Challenges with the increased costs when upgrading RAM and storage. 02:39 *🛠️ M2 Ultra Mac Pro Expandability* - The M2 Ultra Mac Pro's expandability as a major advantage. - Upgraded internal configurations for additional performance capabilities. 03:49 *🏗️ Maximizing Mac Mini Potential* - Bringing the M4 Mac Mini to maximum performance capability through specific upgrades. - Constraints on SSD space and memory due to cost and supply issues. 05:07 *🧪 Testing Methodology and Devices* - Introduction of testing methodology and devices, including M2 Ultra Mac Pro and Threadripper PC. - Explanation of roles and expectations for each device in the benchmarks. 06:30 *📈 Performance Benchmark Results* - Analysis of performance benchmarks including Geekbench and Cinebench results. - M4 Pro showing surprising results against M2 Ultra, particularly in single-core performance. 07:50 *💪 Threadripper Performance* - Highlighting the impressive performance of the Threadripper in multi-core processing tasks. - Comparison of scores indicating its superiority in high-demand parallel processing tasks. 10:07 *🎬 Real-World Use Case: Video Rendering* - Example of rendering 4K video to demonstrate real-world capabilities of M2 Ultra and M4 Pro. - M2 Ultra shows significant speed advantage due to its more powerful GPU. 11:33 *🌐 Everyday Performance and Usability* - Importance of single-core performance for everyday tasks and UI responsiveness. - The M4 Pro excels in standard daily use tasks with high single-core scores. 12:58 *🏆 Choosing Based on Needs* - Discussion on choosing the right machine based on workload needs, not just benchmarks. - Highlights differences in use cases for single-core vs. multi-core strengths among devices. Made with HARPA AI

[@UnlikelyToRemember]

"Overkill is just the right amount of kill" ;)

[@edwardsteinshouer2694]

I pre-ordered the M4 Base model and by the time it arrived I had made the decision to return it and go for the M4 Pro. Never even opened the box and went over to the apple store and returned it for the M4 Pro and even tho I am not a heavy user, the difference from my M2 mini is amazing to say the least. Since the Pro has 512 GB I didn't upgrade it and am using 2 TB of external NVME for storage.

[@ccmmrrnn]

Did you make any upgrades or just stick with the base 12/16/16 M4 Pro and 24GB memory? Do you use it for work? Where do you most notice the difference?

[@LTVoyager]

Excellent job on showing how the real world performance for most of us differs dramatically from specialized benchmarks performance which often only apply to 5% or less of the user base.

[@alealejandroooooo]

Best video I've watched on testing Mac Mini performance (and I have watched a lot...). You explained better than anyone else what I believe is the main concept to understand: single core Vs multi core performance. Then how these cores affect usage and who may benefit from what. Super.

[@cjc363636]

This is so cool. A test that prioritizes the user needs vs. just 'winning' numbers. Thanks for sharing!

[@fab672000]

Maybe another major use case that seems to be missed is DAW (Digital Audio Workstion) workflows, where there are often situations where you run out of cores for real-time audio DSP processing.

[@DaveManDaveDude]

Really enjoyed tonight’s episode, thanks for clarifying that most daily tasks are single cored. Great delivery and information as always :-)

[@m4nc1n1]

Dave, I have the M1 Mini with 16 GB of RAM and 10G NIC. I just picked up the base model M4 Mini. Outside of benchmarks, you would not know which one you are using. So, in real-world use, if somebody has the M1, I would not waste the money on an upgrade to the M4. Especially if you are buying the base model, then you already know you are just doing basic tasks. Great video!

[@StuffYouMightLike]

Load a Logic Pro with 100 or so tracks up on each of those machines and you'll know which is which very quickly.

[@gumbilicious1]

Yeah, but he isn’t doing that. For a daily driver I wouldn’t sweat it, an m1 with 16gb ram would do you just fine and I’d agree it’s not worth the upgrade to m4 in that case I do music production on my m1,I am kinda a lightweight, but the pro chip with 24gb of ram is looking very tempting as a tasty upgrade. For the most part though I doubt I’d notice much of a difference for my normal workload

[@aussiegruber86]

If you are using it as a file server sure. If you are running modelling, video exports etc you can’t rely on slow speeds of Apple.

[@GetVladimir]

Thank you so much for this info and comparison! I was wondering how much noticeable is the difference between M1 and M4 in standard everyday tasks

[@driliagor]

Outside of “benchmarks”? My everyday work is combining LR pants from 21 images I got from DJI drone or Sony camera, timelaps creation and rendering, 4K video editing and exporting. In ALL these M4 just buries M1. If your daily “use” are KZbin videos watching emails and web then yes , stay with M1. Remember what’s “benchmarks” for one is daily work for another.

[@helder4u]

always love your no-nonsense approach🤩

[@kr4bz]

One thing that stands out among the M-series chips is the performance to power ratio. Lots of power and lots of battery life.

[@LarsBjerregaard]

Yes, and since most people use laptops the M-series machines are a minor miracle in that respect. I bought a M1 Macbook Air for my wife's birthday and it's honestly closest to the perfect computer I've ever experienced.

[@PassiveSmoking]

@@LarsBjerregaard I went from an Intel laptop to an M3 Max, and it felt like the scales fell from my eyes. It's unbelievable we accepted laptops as they were after seeing what they can be now!

[@k4everut]

​@LarsBjerregaard I think your statement is irrelevant, given that ALL of the tested machines were desktop PCs (Both Minis and the threadripper). I have a gaming laptop and a Surface Pro that do just what I need them to do as a laptop and tablet. However, if I was one of the 1% of people who edit videos, or I needed a much more powerful computer, the best bang for the buck would be the Threadripper. None of the Macs could compete, as the benchmarks show.

[@Teluric2]

​@@PassiveSmokingi3 6 th gen intel.

[@MrAkaacer]

Laptop is weird. If you're doing video editing you want the fastest to produce more video's. If its just entertainment, web browsing, software development, etc... do you really care about all that power? Your mac is spending most it time waiting for those types tasks. PC laptops are more general purpose than Macs which is what you want in a mobility device.

[@rickbattle5706]

I just picked up a MacBook Pro M3 Max. Got a great deal the day after the M4 was delivered. 14 core CPU and 30 core GPU. I do some M&S work creating 3D color images of Amateur Radio antenna radiation patterns. Love this Mac! Great video and very enlightening.

[@technicolourmyles]

This sounds super interesting! Are there any videos out there for seeing how the M3 Max handles such workflows?

[@silver_crone]

Really clearly explained video, this was great! I got my mini in last week as I switch from PC to mac with nearly the same specs as you used in the test, and mostly watched this for validation that yeah, spending that much was a good idea. :)

[@David_Best]

I find it hard to relate to these particular benchmark tests. My interest is not in video processing, but in 3D CAD modeling performance. I just took delivery of my new Mac Mini with the M4 Pro, 64GB, and 2TB SSD. Whopping $2,800. This week I stress tested it using Fusion 360 against my M2 Macbook Air (24GB, 2TB SSD), and my 2020 iMac with Intel i9 8-core processor, 64GB RAM, Radeon Pro 580X with 8GB of GDDR5. Here are the results for a Fusion 360 complex Sweep (min:sec). iMac i9: 27:47 , M2 Macbook Air 16:46 , M4 Pro Mini. 12:47. Not as much of a difference as I expected.

[@City_Lights]

Thank you Dave, I've had the top spec Mac mini Pro in my cart and been waiting on your review

[@guyloughridge4628]

Outstanding video! Your explanations regarding "everyday use" is compelling, because most of us do not crunch 4K video or do jobs that require parallel processing. The M4 looks compelling, So I learned quite bit today... single core performance gives us a better daily experience.

[@HDEFMAN1]

Appreciate the humour Dave. Also a really good breakdown on these two Mac's. Did not know the Threadripper was such a beast. Hey, I knew it was powerful, but not this powerful.

[@howardwilliams2587]

I ordered my M4 Pro last week to arrive early December. What's shocking is we built out the exact same system! Great minds think alike - at least that's what I'm going with. This will be a significant upgrade when compared to my Z440 with a Xeon E5-2630. Thanks for being hyper active, I thoroughly enjoy the content.

[@DavesGarage]

Indeed, that seemed to be the "sweet spot". If I need more storage, TB4/5 will do!

[@ericbremiller9531]

this was a great video! really enjoyed it! Thanks!!

[@1MinuteFlipDoc]

was hopping for tests showing LLM related performance.

[@alcidem.jaramillo5535]

Great video Dave! Thank you!

[@ansoncall6497]

Haha, hilariously fun. I'm that rare individual who has the Threadripper and the mac. Love both.

[@robertshoop8181]

Yep, same! I don’t have a 7990wx but I have a first gen 2990wx and two of the new M4 minis, one pro and one base model.

[@keoghanwhimsically2268]

I do too, but a Threadripper Pro 7995WX with twin 6000 ADAs? That’s three times the cost of his Mac Pro and likely power draw as well. 😂 Yes, that’s a workstation, but few people are going to put $30K+ into a workstation. I have the previous gen 5975WX with twin A6000s, and even that’s overkill for vast majority of workstation users today.

[@ansoncall6497]

@@keoghanwhimsically2268 🤣

[@kidigus]

This is the best benchmark video I have seen.

[@alexcastas8405]

6:27 just that plot alone showcases the performance and efficiency increase of the m4 soc compared to the ultra, which was only 2 years ago! Just crazy. Great review and info as always Dave!

[@FredHendrix]

Thank you for the spectacularly balanced review and well reasoned analysis

[@ahmu.k6282]

exactly the comparison I've been looking for !

[@WilliamB78]

Superb video. Very informative, with being easily understood for bonus points! Thanks much.

[@jordanaire7329]

I would like to see these do a Solidworks benchmark. Specifically single core processing for large assembly drawings.

[@larsgotfredsen6068]

Great review. For completeness price of each system would have been nice. A threadripper pro 96 core cpu is approximately 10.000$ 😅

[@buzuli]

The Final Cut comparison is where the choice of the M4 Pro vs. M4 Max chip is really apparent. A Max doubles the video transcoder count of a Pro. An Ultra doubles that again. Having 4x the number of transcoders in the media engine explains the nearly 1/4 time of the M2 ultra over the M4 Pro. I would expect an M4 Max to close this gap to well under half.

[@shanebradfield]

i have never seen ur channel before today but wow u nailed this review 10/10 earned a sub and a share to my mates.

[@GrzegorzDerebeckiXan]

Excellent video. Nice that You compared with 7995WX!

[@fredflickinger643]

Thanks for the perspective!

[@h069401]

Bravo Dave, I love your "women & children first" approach to family automotive resource allocation. Respect.

[@mr_b_hhc]

Great video with clear and concise explanations, thank you.

[@tomtrottier8135]

I enjoy your videos, even ones with dream machines few can afford. But since the M4 pro is only 7% faster than the plain M4, the $600 machine may best your (non-M4) beasts on Speedometer, and therefore on performance ordinary users notice. You gave some prices, but what does your threadripper cost? Thanks again

[@maxwellsmart3156]

Speedometer 3 measures javascript performance and is browser dependent. It just happens to be single threaded. I got a 38.4 on a 9900x with 5200 MT/s memory using Chrome, it was surprising that it was that close.

[@LarsBjerregaard]

Fair, balanced and articulate verdict Dave, completely agree.

[@ronm6585]

Thanks Dave.

[@Rhythmattica]

Wow... i love this channel... im a layman in many terms, but love the "this is how it is" delivery ..No F'in around....... Keep on keeping on.....

[@milosp]

How about then single thread Intel CPU with lower core count but a higher clock then the TR for the last benchmark?

[@garanceadrosehn9691]

Very nice to see all these stats and various benchmarks. I'll have to spend some time running some of the implementations in the Github Primes project as mentioned 8:54 into the video. _(It happens that I wrote a primes-sieve program of my own several years ago, written in a language called Crystal. I expect that wouldn't complete with versions in the Github primes project though!)._ I reached many of the same conclusions when ordering my M4 Mac Mini as you went with here. I felt that 1-TB SSD was the minimum, and as much as I like Apple I couldn't convince myself to pay their prices for SSD upgrades over 1-TB. I was pretty tempted to go for 64-GB memory, but when I selected that it would have delayed delivery by a few weeks so I stuck with 48-GB. And 10-GB ethernet because, "Why Not?". The only difference is that I stuck with the 10-core model instead of the 12-core model. As it is I'm sure this will be massive overkill for everything I do on a Mac at home.

[@DavesGarage]

Check out the Crystal solution! github.com/PlummersSoftwareLLC/Primes/blob/drag-race/PrimeCrystal/solution_1/primes.cr

[@SaHaRaSquad]

Yeah for SSD I'd just buy a fast USB drive, you can get 2TB for less than a single Apple storage upgrade.

[@KeithPhillips]

Wow this couldn't have come at a better time. I'm literally looking at the M4 MacBooks as this was posted. 😂

[@AFITgrad86]

Thanks ... very informative. I may upgrade my M2 Ultra at some point to an M4 for as you say the single core performance boost since much of my work is writing and web browsing with some video (mostly ripping) work.

[@Wrackey]

Getting you to upgrade the base model is exactly the point with apple. For reference, 64GB of CL30 DDR5 6000 can be had for under $200 on PC

[@Dirtyharry70585]

Dave what about the m chips that all have compromising machine core embedded code? Apples fix was bypassing using software? Did they ever revise the CPU’s?

[@aaronbielish7087]

Can we do a comparison of units of efficiency/watts used between the three?

[@andrewparker8636]

What would have been nice to see is the M4 base model vs the M4 upgraded on the "daily tasks" style tests. I'd like to know if there's any reason to upgrade if your workload is fairly mundane

[@altosack]

If one doesn’t know if they need to upgrade beyond the base model M4, then they don’t.

[@kplkasteel]

A true and honest speed test comparison test that just doesn't emphasize on Apple bling bling to good to be true stories by all the Apple fan boys.

[@RichardGetzPhotography]

5:07 did you purposely not mention the price?

[@EnronnSierra]

Still can't find a reason to upgrade from my 2020 M1 MBP. I do home to replace my Windows laptop with potentially a Mac Mini or Studio when Windows stops getting updates. Its running Windows 11 now, with 32 GBs of RAM, hooked up with three 24 inch monitors. So, its definitely meeting my needs.

[@acasualviewer5861]

I really need that 10K Threadripper + 15k dual A6000s to run my minesweeper workload. Imagine how quickly I could play with that!

[@JohnCorrUK]

Another brilliant video Dave … you are helping me define my 2025 upgrades on OSX and Windows

[@JSyntax]

Best review of any technology Gadget and this guy is loaded

[@X.MillennialResponder.X]

I don’t recall, you mentioning the thread ripper power consumption versus the Mac overall I think what for what the efficiency is so much better on a Mac nowadays it’s gonna be nuts to see if they do release the M4 extreme to see how that then compares what to watch and performance of that of a thread ripper

[@RedShiftedDollar]

The m4 single core perf makes the iMac a legitimately good choice for a total computer refresh. The iMac is geared towards light use general information consumption users and the m4 punches above its pay grade in the snappiness arena. Plus the screen is very high quality in terms of color accuracy and resolution which is perfect for media consumption.

[@Cybernetic_Systems]

I got the exact same M4 Pro Mini, but with 64Gb ram, it’s a tiny beast! I replaced an i5 mini and boy is it an upgrade!

[@marklong8608]

Would love to see what the comparable mini-PC is for running windows and if the base M4-mini is a better option if you need a windows machine. (some of the software I need is only available for windows).

[@Biaanca5036]

I wish I could ask Chuck Sommerville how the Visual Studio Spy utility and the Spy in Windows Chip's Challenge ended up having the same icon

[@oliverpigram7054]

I got the maxed out M4 Pro Mac Mini with 48Gb ram for 4k video editing! It's a beast.

[@MVHens]

I was told that connecting an old iMAC as a monitor to a new M4 is not possible. Is there some technical explanation why this does not work?

[@alwe_st]

Great video, thank you. I'm going wait out for the M4 Ultra or maybe the Extreme.

[@DonNoll]

Great info Dave.

[@SeanLynchXY]

Thank you, Dave, I was curious as to how the M4 Pro would compare to enterprise class workstations and was happy to learn about single core performance v. workload considerations. Other interesting metrics might be gained by normalizing the score on each test by dividing the test results of each system by the lowest score on each test. Then assigning points based on the resultant value and summing the points each test for each system. The sum of the points could then be used to evaluate performance points per dollar or points per watt or even use weighted individual test scores based upon intended use case for each system.

[@Pracedru]

Wow.. That single core speed is incredible. I am kinda blown away by that.

[@dna598]

great video. The M4 of M4 reviews.

[@DosMan94]

Finally a real test, rendering a file , its takes 3 Min for the M4 and 2 Seconds for the faster machine ..... everyone gets that :)

[@kdeuler]

Great essay as always. Thx.

[@machdaddy6451]

Maybe you could talk about the graphics of the "Sphere" in Las Vegas?

[@barryinkpen6026]

Very enlightening. Thanks !

[@SAerror1]

How much memory does the prime sieve test code use? Does it fit entirely in the L1 or L2 cache on these modern CPUs? One of the coolest things about the M series processors for me is the 800GB/s main memory bandwith but I guess that didn't help here! The scores looks downright pedestrian compared to the 7995wx, with a gap that big it might be comparable on compute per watt too?

[@loganmedia1142]

I've found for both user interface responsiveness and multicore tasks my Windows laptop and work Macbook Pro are essentially the same. Other than better battery life the Macbook Pro's main distinguishing feature is that it cost 3-4 times as much.

[@MatrixRoland]

Your Threadripper desktop looks sweet!

[@pauljones9150]

Actually a great show of single vs multi core

[@khoifoto]

The M4 is a very capable chip. I actually sell my M2 Ultra Studio and get the M4 Max because I need to travel soon and I'm happy to see M4 max is just as fast as the M2 ultra in many cases.

[@AjinkyaSChavan]

for me it's a weird IDE issue at this moment, Xcode cannot run on Windows and Visual Studio cannot properly run on Mac 😂 for my development needs of Swift and C#

[@for2utube]

Thanks. I was saying "what about something like openssl speed -evp aes-256-cbc -multi 32" when you brought up your primes benchmark.

[@HISPEKK]

Could you do the test with topaz please, not sure if that uses single core, thanks for great video.

[@94040]

Thank you 😊

[@lxndrlbr]

A crucial piece of information to contextualize the comparison is the price of the Threadripper PC with dual nVidia 6000 GPU ... I am not sure it was far below the M4 price 😅

[@LuvADuc]

Thank you for the content.

[@D.u.d.e.r]

Thx for the well made vid!👍Only base M4 Mac Mini with $600 price tag or less is a great value and indeed Apple made a great choice to offer it now in this economy. However as u have pretty well pointed out, once u start upgrading it u realize how ridiculous Apple's "upgrading" truly is, so I guess most people will stay away from it. At least there r ways to upgrade modular SSD storage either by the repair/upgrade shops/DIY or later on by the certified storage upgrades (OWC etc.). Apple has definitely captured some new hearts & minds with it and hopefully have realized what is the recipe for expanding Apple user base. For me PC has one great benefit which I will always value from multiple points of view and that's modularity especially with the desktop FF. Only thing Apple needs to do with all of their Macs in the future is to make them all SSD upgradeable in similar way as Mac Mini both by themselves and through their certified partners. Users in general will accept more Mx chip + RAM fixed and not possible of upgrade if SSD storage is upgradeable. Also decision of the "design" team or whoever came up with it, to put power button on the bottom of it was a quite detached from reality. Having it on the back would be the best solution from both design and functional side. It's probably the same person/team which opt for stupid camera notch on the MBP or put charging plug on the bottom of Apple "magic" mouse👎Some Apple design choices simply don't make any sense...

[@kevin-e5h5t]

"...and too much is just about right". As a 8K Blackmagic Raw Editor, using DaVinci Resolve, that is true.

[@halsterhof]

Thanks for an excellent video. So if you configured the Threadripper PC with same amount of RAM and storage as the Mac Pro has here, and only one GPU, how would the cost compare?

[@VforVictorYT]

I am interested in the break point, what is the heaviest single core task and at what speed doesn't it matter how fast it is?

[@satchell78]

The size alone almost makes it worth looking into doing Win in a VM and ditching my pc. Dave, what would the trade off be?

[@dtsdigitalden5023]

Dave "Harryhausen" Plummer. Loved the intro!

[@Lemurcoon]

Gonna need a spec breakdown of that monster of an AMD machine, please and thank you

[@alliejr]

Excellent!!! However when testing video render exports (10:30) aren’t you testing your ultrafast RAID 0 disk as much as CPU/GPU? I suspect a multi-gigabyte export process is as much IO bound as anything. Am I wrong?

[@slipoch6635]

I'd be curious to see how resolve would go on all 3 as it is the industry standard for denoising and BMRaw is one of the more accessible high fidelity high res file formats (prores cannot do some of the higher resolutions). That being said the M series has a bunch of lookup chips for various video codecs so it may not be a completely apples -> apples comparison, but definitely interesting as traditionally I have found resolve works better on PC even though they primarily target Apple these days.