How good are the Apple M4 chips when rendering in Blender?

[@madebyjonny7637]

finally someone has done a real test, thank you for this vid

[@RobbieTilton]

no problem. thanks for watching!

[@தமிழோன்]

Not only that, he got straight to the point... he showed the results in 2.35 minutes! Pretty impressive given that most KZbinrs love to ramble a lot.

[@DarrinNoNAME]

Yeah it’s nice to see someone doing a real render test with these m4 chips. I really would like to see animation fps speed in the viewport.

[@axi0matic]

@@RobbieTilton Did these tests use Optix for the Nvidia cards, or just CUDA? What CPU does the PC have, and how much RAM?

[@RobbieTilton]

@@axi0matic CUDA. Just added PC CPU/Ram to description

[@joshuaschroeder2421]

I loved how you were straight to the point and delivered good results. You earned a sub.. Keep it up

[@RobbieTilton]

thx brotha

[@jjmcook]

Agreed and subbed.

[@thanhclips]

The best part of this video. I didn't have to wait for an introduction of blah blah blah for over 5 min.

[@HeangKoing]

@@RobbieTilton Great vid showing real tests relative to others. No faff, great vid!

[@Seb00lean]

Straight to the point, no unneeded fluff. Subbed!

[@artofjhill]

awesome thanks for doing this. Waiting for that Studio with Ultra 🙏

[@KellyWu04]

I bought an M4 Pro MBP 14” for software development and occasional video editing and other creative work. It’s a beast.

[@ronaldbell7429]

I've got an M1 Max MBP. The M4 is seriously tempting.

[@jarhead4657]

@@ronaldbell7429 replaced my M1Max Quran an M4Max. Pretty much twice as fast at everything. Plus many features the Maxlaptop doesn’t have.

[@greentokyo]

Very impressive results! Thanks for testing this!

[@RobbieTilton]

no prob. thx for watching

[@JFinns]

Not impressive at all, way behind RTX cards in a real test with OptiX not disabled.

[@heyzoltan_cgi]

M4 Max chips are super impressive, replaced my base M1 with the binned M4 Max and it renders on average 17 times faster in Blender. Comparing it to the 4080, the fact that a tiny laptop GPU can reach around 50% of the performance of the strongest desktop GPU while using a fraction of energy is ridiculous.

[@markminch1906]

It all comes down to the fact that the GPU for the mac is on the same chip as the ram, and cpu. The amount of energy and latency that comes about sending signals across centimeters of wires vs nano meters of wires is huge. The M4 series memory bandwidth makes even ddr5 ram seem slow in comparison for this reason and the same goes for cutting down delays of sending data from the cpu to the gpu

[@sveilien]

@@heyzoltan_cgi I need to bring a Blender project on an portable with Blender portable to Best Buy and run a benchmark on their floor model M4 Max to compare to my M2 Max, haha.

[@peraman2022]

@@markminch1906Apple’s game of power-efficiency is definitely paying off.

[@utubekullanicisi]

The M4 Max isn't tiny, the overall SoC size (including the CPU GPU etc.) is bigger than the 4080 Super and probably close to the 4090's 609mm2. If you're comparing the cooling systems it might be.

[@peraman2022]

@ that’s what I’m comparing, the power draw.

[@mdumabhena2024]

This is the review i have been waiting for. Thank you very much

[@josefludvikbohm5390]

Thank you! Would great also to see ram / vram / unified memory usage!

[@sloanNYC]

Very interesting comparison, crazy to see an SoC keeping up so well. Your charts are a bit wonky though. The Monster Under the Bed one for example the 4080 and the M4 Max look nearly identical because you compare it to the slowest possible, when really the 4080 is 43% faster than the M4 Max finishing in 22 sec vs 34. That's 12 seconds or ⅓ less time. Essentially including the very slow M4 (so few GPU cores) as your baseline makes everything relatively much closer together because the gap between M4 and everything is so huge, it makes all the other gaps seem small.

[@kepa7983]

Without OPTIX in 4080 Super? Extrange comparison.

[@movescalerotate6715]

Cool test, thanks for uploading! I just purchased a 16" M4 Max Macbook Pro. Monster under the bed scene, Macbook gets it done in 25 secs, while 4090 is 16 secs. Not bad at all.

[@GlobalWave1]

Next year they should add 10 more GPU cores to the M5 max since they have to compete with the new 50 series. I wonder how the M4 ultra will fair.

[@wisdomyaw03]

Which of the 4090? mobile or desktop?

[@BlueHawk80]

@@wisdomyaw03 Even vs the 4090 mobile this would be impressive, considering the power draw.

[@seriessplayer62747]

@@wisdomyaw03must be mobile, because the 4090 mobile was much faster than the m3 max. It would be stupid to buy a m4 max when you have a 4090 desktop

[@moozoo2589]

@@seriessplayer62747 not so stupid considering 128GB of unified memory, 4090 has only 24GB. Right now you can load huge LLMs into a Macbook and run them fast, you can't do the same with 4090.

[@RetroSoulProductions]

Contemplating on buying a M4 Max MacBook Pro or waiting for the M4 Max Mac Studio. I’m running Blender, Unreal Engine, DaVinci Resolve and FL Studio and much more.

[@GatoPaint]

I have the 40 core M3 Max and this M4 Max with 32 cores performs the same, seems like the 40 Core M4 Max might have even better performance

[@johncho8766]

Great video. I'd love something to show viewport performance while modeling, texturing and lighting.

[@guillaumemuller-greven1425]

More reviews should be like this. Straight to the point. Immediately subscribed.

[@CorentinHarbelot]

Now, we are talking! Hope the M4 Ultra is coming one day..

[@theorangeslice9678]

I intended to build a rendering PC with an RTX 4070 and was eying Apple's newest stuff to see if it'd be worth it to just go with that instead. And it 100% is. I've been watching Apple's hardware improvements since they started using their own silicon in their products, and I was waiting for something like this to happen. It's mind-boggling how much they're improved in the past 6 years, going from being a product that you got for just the logo to a legitimate threat to Windows PCs for productivity purposes.

[@joerama]

According to most rumors (search for Mark Gurman), the Mac Studio (and Apple Silicon M4 Ultra) ETA is mid 2025, with Mac Pro (tower) shortly after. The M5 chip design is rumored to be complete, so the M5 line up will start rolling late 2025. Ultra is always last to show due to its cumulative complexity. So don’t expect M5 Ultra until late 2026 (or so). The other rumor which has never manifested is the Apple Silicon “Extreme” which is 2xUltra. Perhaps that will become reality (Mac Pro tower) with the M4 round of Apple Silicon (2H 2025)? TY for the graphics comparisons.

[@chidorirasenganz]

Thanks for commenting this so I didn’t have to 😂

[@arkabhowmik7832]

@joerama Apple will release the M4 Ultra by mid 2025. But M5 Ultra probably won’t be a thing and they will most likely directly jump to M6 Ultra. Honestly, that’s the way to go. Very few people actually upgrade these high end desktops frequently

[@chidorirasenganz]

@ nah M3 was only skipped due to the N3E node yield issues. From now on it doesn’t make sense to skip generations

[@TechGameDev]

what is this mac 😂wait it will consume 4x the M4 max the worst for 2026 and the price of $10,000 (apple gave us gpu support on the mac pro it's possible) around 400w

[@chidorirasenganz]

@@TechGameDev The Ultra only consumes about a 100w so an Extreme would simply be 200w. Nowhere near 400w

[@AshTheZombie3D]

Hell yeah, head to head with details, you rock.

[@christopheralan]

Nice work, my OG M1 iMac with 8GB took quite a bit longer 😝 35 min for the second one. One minor quibble is your chart graphics makes it feel like the bottom one is fastest (without reading labels) because it has the shorter bar....

[@caseyleedom6771]

In the Blender relative performance graph starting at 0:39 mm:ss, the relative sizes of the bars seems weird. For instance, the NVIDIA 3080 is almost half the speed of the 4080 and yet it's bar isn't half the size ...

[@sveilien]

Curious; I just did the Monster one at 1024 samples on my M2 Max MBP. It did it in 1:15, faster the the M2 Ultra on this chart. Was your rendered as-is from the original .blend?

[@nintendoever]

I did the same test on my Macbook Pro m2 Max - 38-core, and it took 1:19.

[@sveilien]

@@nintendoever I just did the other one and it was 5:56, so longer that the Ultra.

[@nintendoever]

I also did the Lone Monk test, and it took 06:12 on the same Macbook Pro M2 Max 38 core. But raw GPU performance (which impact games) on M2 Max is great, 25% faster than M4 Pro.

[@sloanNYC]

@@nintendoever It would be great if we could just get an Apple GPU card of some sort to really crank on cores...

[@nintendoever]

@@sloanNYC I agree and I it is perfectly viable. Especially now, with thunderbolt 5.

[@tomcollins6989]

🙏. Finally the test I needed. Thank so much dude.

[@christopher3d475]

Was Blender configured in System to use Metal (both CPU and GPU) and the Kernel set to Full?

[@RobbieTilton]

Configured to use metal GPU / kernel set to full. I did not check the second cpu option.

[@mariuswickli3206]

CPU and GPU are (at least in all scenes i have ever tested them) way slower than just the GPU alone (M1 Max, 64GB Ram)

[@chidorirasenganz]

@@mariuswickli3206Yeah I can confirm this on my M1 Pro as well

[@nintendoever]

How do you set kernel to full?

[@galder3112]

Yes, I checked it now and thats right! On M2 Max 32GB Mac Studio with checked GPU + CPU in system settings I had 40sec in Cycles. With only GPU checked the result was 32 sec :O. Its huge man, wow... its additional 20% faster with no effort. I mean, I tested my own project ;) Blender should add a note in system settings in Blender

[@Concodroid]

When I rendered the monster scene with nothing changed on my 4080 super it had a time of 15.17 seconds, not 22; are you using OptiX or CUDA under blender preferences - system?

[@kepa7983]

CUDA

[@mishanya5443]

Явно CUDA использовалась, тогда бы не получилось сделать подобный ролик, ибо отрыв у M4 MAX от 4080 super был бы такой же, как у 3080 с 4080 super почти, пару секунд M4 MAX выйгрывал бы у 3080

[@itiswhatitis-yes]

Incredibble video man!

[@mikey7326]

Those results look good but... how does it compare when using optix rendering?

[@pfolger]

16” MBP M3Pro 18 Core GPU 36GB RAM. Settings GPU only, Kernel = Full, MetalRT = On, Denoise = On, GPU = On. Multiple renders, average = Monster = 1:08. Lone Monk = 5:37. Also waiting for the Mac Studio with 60+ GPU Cores. Rendering Lone Monk with a second monitor (4K at 4K) and laptop screen on, added (avg) 18 seconds to the render. FYI. Thanks for the truly useful video!

[@utubekullanicisi]

Did you use OptiX for the Nvidia GPUs? Not using it will significantly hold the Nvidia cards back as the cards won't use their ray tracing hardware to accelerate 3D rendering without it.

[@kepa7983]

CUDA

[@NomadicBloke1]

Thanks Robbie for being direct to the point and doesn't even bother to make it long in order to gain profit. I'm going for the base model at 24gb as an upgrade to my 2019 5k imac. I use blender for basic modeling of jewelries and uses Light Tracer Render.

[@raviranjan554]

Can we get a video on after effects comparing m4 variants with desktop cpu & gpu combined for both of system.

[@RobbieTilton]

sorry brother - i'm not an after effects person.

[@arkabhowmik7832]

After Effects runs very well on Apple Silicon. I've compared my base M1 Pro against a i7 12700K, RTX 3060, 32 GB Ram build which came to a similar price since I got a great deal on the MacBook Pro. My M1 Pro came out on top by a significant margin almost always!

[@noisetin]

Thank you, I was waiting for this! So basically almost 4080, which is kinda impressive how with each generation of AS they're closing the gap. With M5/M6 they should be on par with the fastest RTX's.

[@GlobalWave1]

Well the M4 studio will hopefully be out next year and at least be on par with the RTX. They should honestly add 10 more cores to the Max or give the option.

[@noisetin]

@@GlobalWave1 M4 Studio will be double the Max, so 32c CPU and 80c GPU max variant.

[@chidorirasenganz]

@@noisetinThat would be the Ultra variant which would be a lot more expensive. Also we might get the rumored “Extreme” Chip this time which would be two Ultras fused together

[@ap-zm3bl]

the graphs are not proportional … the m4 is 50 % slower than the 4080 in the monster scene

[@noisetin]

@@ap-zm3bl Oh indeed, I was tricked! You're right

[@numosis3627]

Thanks Robbie, this is super helpful!! I’m in a similar situation. I’m most likely going to switch from a PC to a Mac Studio M4 Ultra once it’s released. Since I render most of my longer sequences in the cloud anyway, this might be the right move.

[@asafblasbergvideographer]

Great video Robbie! And I'm sure you'd get a slightly higher score if you tested the unbinned M4 MAX variant 16-core CPU, 40-core GPU. Now that the 5090 was announced, Apple will have to catch up again in its next chip release.

[@coloniapobre]

These are going to get insane in next couple years….wow

[@robertkamer1820]

I just ordered an M4 Max last week, for After Effects and Blender.

[@zeuskyst]

This is very interesting. I wonder what the M4 ultra might look like on a blender benchmark. Thanks for sharing.

[@Kullervo456]

Pardon my ignorance. Is m4 and max version good for high poly sculpting without view port delay?

[@cmscss]

Thanks for this, just the info I needed.

[@d3pixels]

Fast and clear … Thank you for the info. Good video.

[@JasonEbeyer]

Literally the exact video I've been waiting for. Thank you for this! I think if they release a M4 Ultra in a Mac Studio I'll be making the switch from PC.

[@LuboBachev]

I'm a 2d motion designer and I don't do 3D, but i was hoping the M4 Max would be enough for me to start learning on it, thank you for the video, because it made me confident that it will be plenty :)

[@gabbo13]

Simple and direct! Loved it!

[@scottdpeterstv]

I think this would be more interesting using laptop dGPUs instead of desktop, but cool to see. Looks like with M4 and beyond Apple has finally caught up to nvidia with a power efficiency advantage. Very exciting

[@koto9x]

short simple and to the point. subscribed

[@LoveLineStories]

Dear , Should i buy this m4 base mini model for blender .. i learn documentary like fern , edits like imperial , plz tell me this base is enough ?

[@gabriellasatori7600]

Do you think the normal Pro Max version is enough for C4D or is it worth to switch bigger unified memory version ?

[@echowhiskey_gaming8284]

lets see the ultra chip performance once the studio gets released

[@the9thlion]

what about m4 pro on mac mini? not the base m4. curious

[@dezem7]

For portability and battery, the MacBook Pro m4 max seems like a good choice over a ball burning windows laptop until the PC arm versions catch up... ?

[@San_OO7]

You earned a sub bro 🎉

[@FascinateFelix]

Wait for the Studio M4 Ultra it it'll be spring most likely. It'll be likely what I'll aim for.

[@TheAiGuy100]

Thanks for this! Just wondering, was MetalRT enabled?

[@AngryApple]

Was OptiX used?

[@RobbieTilton]

@@AngryApple CUDA

[@AngryApple]

@RobbieTilton Re Test with OptiX please would be interesting to see. Because why not using OptiX?

[@realhamza2001]

optix is nothing but a fast denoiser it doesnt actually matter when it comes to the final render speed

[@AngryApple]

@@realhamza2001 thats not true at all. Its an acceleration structure. OptiX Denoise is something different. CUDA isnt using the RTX cores, OptiX is. The Denoiser is using the Tensor Cores if available

[@PKperformanceEU]

@@AngryApplestop spreading bs Accept that apple is coming for Nvidia there s nothing u can do about it sheep

[@SerjLimitless]

Been using an M3 max for 10 month in C4D & Houdini & Redshift - extremely happy with it Multiplied with portability, display, sound and all other Mac features makes it an absolute champ for me

[@Fakepilot]

I used to travel around nomad style, working in different countries with a Mac laptop doing Flash animations. Now I do 3D, with Houdini and After Effects. Considering buying the M4 Max, that it will be enough to do good preview renders and then send full scenes to a render farm. Maybe I could do that nomad style life again? ;-)

[@lilianwhite3449]

Did u use optix?

[@fluxcapacitor]

When purchasing a new Mac mini M4 Pro, with Blender GPU rendering in mind for complex scenes (i.e. lot of quads or triangles), would you put a few more dollars either on the 2nd processor (going from 16-core to 20-core GPU), or on the RAM (going from 24 to 48GB)? It is my understanding that the 20-core GPU would be faster, but if the scene eats too much RAM for that 24GB machine, the rendering performance will collapse, which would not happen on the 48GB machine (except in insanely complex scenes, but that's not an option in this example).

[@art-thou-gomeo]

Love the no-nonsense approach - instant subscribe! Quick request: any chance you could compare Metal vs NVIDIA DXR real-time ray tracing? GravityMark should let you test this - it's a comparison we haven't seen yet!

[@3DJay429]

They’re definitely impressive! I’ve been using an M2 for a while, but recently switched back to Windows. While the Mac performs great in benchmarks, I found it sometimes lags in real-life scenarios. For example, when I have multiple browsers open or visit pages heavy with banners, it noticeably slows down. I also missed the customization options on Windows. I recently got a 42-inch OLED monitor, and it works seamlessly with Windows. With my Mac, finding a good monitor setup was always a challenge.

[@Wayne-sn6qy]

It's also interesting if we could also know how much time M3 Max needs to take in the same scenario to get to know how much improvement from the ray tracing core.

[@matthewsykes2646]

Very informative video, thank you! I think the most impressive thing is not the ultimate performace of the M4 Max (we are also a couple of months away from the 5000 series from Nvidia) but how efficient it is and how little power it uses to achieve such results. In the lifetime of the machine i think the amount of energy you save while working is huge compared to a desktop or even a Windows laptop with similar performace. I'm very tempted to buy a mac mini with the M4 pro for 3D since i was looking for a Windows mini PC but there's none with the same amount of efficiency and performace, unfortunally. Thanks again!

[@MikeyOfLegend]

How does the M3 Max compare?

[@galenbeals3538]

Great video. Thanks! Just a little FYI: current Mac rumors point to a June (Apple Developer Conference) release for the M4 Ultra Mac Studio. They could surprise us though but I wouldn’t count on it.

[@paolaanimator]

Thank you so much for covering this! I have been wondering about MacBook Pro M4 Max chip being the perfect portable machine for 3D rendering/animation, video editing and coding. I love the battery life and the sturdiness of this laptop. I don’t even have a MacBook Pro, just an underpowered intel MacBook Air. If I ever save up enough, one day I’ll get myself the M4 MacBook Pro. ❤

[@adziak]

Quick & informative. Like deserved!

[@fotiosrappos3997]

did you do the test on macbook pro 14inch or 16inch? Also where the windows pc desktops or laptops?

[@guoyuanasia2]

How did you get all the details on every model?! Good work!

[@mariuspuiu9555]

Is Optix enabled for the Nvidia renders?

[@605dave]

The rumors are the the M4 max and ultra studios will be out next spring

[@joserioslopez9515]

For the renders the results are impressive! But for the generalist 3D machine the Mac Mini M4 Pro are good? I need a new machine and thinking in the M4Pro Mac Mini...

[@JasonCalvertRI]

I use MAC for my 9-5 job. Our whole design ecosystem is MAC but I do 3D modeling & sculpting as a hobby. My iMac is 5yo now and running super slow so I've been looking at the Mac Mini M4 Pro. Its probably the best bet for what I do professionally and as a hobby. Thanks for this video. Stats look like double the time as a high-end PC but way faster than what I'm using now.

[@JoelAustinMack]

I really think we should consider that the M4 can do this unplugged. This is the biggest reason I am switching: I am no longer shackled to a 3-pong outlet and don't have to suffer huge frame drops like I have with my 4080 laptop unplugged.

[@DaniSiem]

Thank you. Your perspective answered and touched on some unanswered points I had in my head. and by what you briefly described as your background, I assume you know know. Nice video. New sub

[@zachwarner7164]

if i have a gaming rig with a strix 4090, should i keep the m4 base or the m4 pro base? I feel like if i edit on the m4 pro, then just unplug and plug into the 4090 for fat deliveries that would be optimal...?

[@marco.lino.27]

Brooo wtf the M4 max is an absolute beast. Cannot wait for Mac Studio

[@saarrrk]

Would love to see a comparison for a scene where 4080 is nearing its Vram limit

[@TechGameDev]

you are not a 3d artist the ram is loaded mainly in the processor ram and a little of the vram and if you are a real artist you will optimize the scene during modeling by masking the objects no interest if a mac 40 gb of vram with little performance it will only be buggy unusable (bottlneck gpu)

[@ridetotheforest]

​@@TechGameDevIf you are working in a game company optimizing is a later task. You need the VRAM IF you are a '3d artist' and delivering the assets in a shorter time period.

[@AngryApple]

@@TechGameDev Big Scenes even split is something that can happen and need a lot of VRAM youre using Proxies for the scene creation state but you will never render the scene if you have not enough memory. The immense disney island demo is a testament to this. Also a benefit of the Apple Silicon is the unified memory, scene startup is significantly faster, doesnt matter if the rendering on Nvidia Cards is faster if the Building step takes so much longer. We have a small render cluster with 1x4090 and 2x4060ti 16GB sometimes we just skip the render cluster because the M2 Max is faster in giving us a result because of the scene building time. Its actually crazy and never thought about it before until I saw it in action. You simply dont have to move data around anymore both the CPU and the GPU have access to it. If a game would optimise for this it would be a extreme boost in performance

[@shaharyar.shaukat]

@@TechGameDev All I wanted to know was if it could load up scenes that take around that vram, I use a 3090 for work and personal use and sometimes have to hit close to its 24gb Vram limit lol. Also a lot of the time iterating in the render preview. Optimization is not my concern as I do concept design and deliverables are fast and have to regularly use UDIMs and stuff such that the textures eat up way more vram than any polygons lol.

[@TheNoiseGOAT]

Are the Nvidia GPUs using optix in these tests ?

[@fotiosrappos3997]

did you do those tests in a 14inch macbook pro or 16 inch?

[@jaketheman7402]

Hi do you think it's necessary for the m4 pro to have 48gb when doing more complex work on blender? Especially animating?

[@oldtimer7635]

Was that Mac Mini the base model?

[@gerardofaustin8156]

I’ve been waiting to switch to Apple Silicon since the M1, but the only thing that was stopping me was the waitress and performance as an architectural engineer when I do my work I need tracing and I also need hardware that will support structural analysis and have the power that I would get from a PC this year is looking quite enticing and I might just buy one. Or hold out for an M5.

[@JUSTINBUISSON]

Simple video, only facts. Thanks.

[@smepable]

This is mindblowing! 😮 I am astonished how much every generation of processors still advances so much over the last one! That Macs will ever beat a high end graphics card of the last Generation is groundbreaking

[@Meetdesigns]

Great Video, straight simple and real world test. Love it ❤

[@Aris0281]

I do archviz in 3ds max with Vray GPU. I would really prefer switching to MacBook pro but there's no 3DS Max for it unfortunately

[@dezem7]

I used 3ds max and realax sgi back in the 90's whilst studying, When I went into the industry it was Maya, and now Blender seems to be taking over... Have you looked into blender at all for your line of work?

[@MobyMotion]

Thanks :D

[@sdshowz]

Thank you so much for this !!

[@vishwsant]

Short and to the point. @1:44 Me too in the same predicament. Can't wait for the Mac Studio to release and hope they keep the same price. Btw, I see that you've used base versions of M4 Pro and MAX. Can you compare them to the top versions if you have them?

[@detore]

3 questions. 1: Can you list the exact settings in blender for the renders you performed? 2: How are you calculating those graphs, they look way off a standard graph with those number of second values in the standard (lower is better) configuration? 3: Why was it interesting that the 10 gpu core m4, was being outperformed by the 60 gpu core m2 ultra? I would think the opposite would be interesting, this results you showed would be expected?

[@sskdev5116]

Thank you!!! I was looking for something like this. Thanks

[@bushgreen260]

*Were they using hardware ray tracing?*

[@RobbieTilton]

yes

[@secondskins-nl]

@@RobbieTilton No, not if you used CUDA instead of Optix

[@mrk-ism]

Interested in how it compares to m3 max

[@SmoStory]

Thanks! What are the specs of the MacBooks? Like how many Core CPU, Core GPU, and how much GB Unified Memory? It makes a huge price difference.

[@RobbieTilton]

Just added in description

[@sleepyelk5955]

thanks for the real test! can you say something about the noise and/or coil whine?

[@karimmusa7644]

what's the music track in the video ?

[@RobbieTilton]

KZbin Copyright Free Music Library: Videodrome - National Sweetheart

[@kinkaidi]

Hi, I'm considering betweet M1 Max 64 GB (used) and the M4 base model due to the similar pricing in my country. Which one is better for 3D and Game Development?

[@secondskins-nl]

I'm not sure why you consider those since rationally you should consider what PC you should buy. Uglier but faster and well, most games are not really for a Mac so in the end you'll be behind a PC anyway. As a bonus it's also much faster if you put the money you save into a better GPU.

[@adm-k8r]

Thank you for your video. I am wondering where on your chart would M1 MAX be.

[@thenout]

Thanks. It's impressive, yes. But those differences in render time are humongous. If you render sequences, those 30secs difference end the conversation.

[@sly_band4459]

Hi thank you for such a informative video I’d love to see what it would be with eeve rendering

[@lukestrat93]

Optix for RTX and Metal for M chips, right?

[@kepa7983]

No, he use CUDA