Noise Reduction & Speed Warp - DaVinci Resolve + M1, M1Pro & M1Max

How does the M1 from last year compare to M1 Pro & M1 Max when it comes to Noise Reduction and Speed Warp?
Join our FREE crash course...the FASTEST & EASIEST way to learn Resolve:
filmsimplified...

[@LearnColorGrading]

Join our FREE crash course...the FASTEST & EASIEST way to learn Resolve: filmsimplified.com/p/davinci-resolve-crash-course

[@DaveDugdaleColorado]

Thanks for sending all the time to do these tests for us. Very helpful.

[@maxm-corp]

Very interesting. Unlike all the tests that test render speed. If I use 3 or 4 days to edit a video, it is not really that important if it renders in 5 or 10 minutes in the end. What matters is the smooth workflow in the 3 to 4 days of editing. And good optimization is where it is at.

[@jaysnappsmediallc]

This 👆🏾

[@Mr326]

Yes!!! Like I’m ready to take a break when I render anyway

[@nerfbutt]

Exactly!

[@Clickumentary]

Yes what he said. Playback and realtime performer during creative process is all that matters.

[@jasonnowwhat5041]

Great comment here. So much is made of export and render times…I couldn’t care less. Once I’m done I’m reaching g for a sandwich. What I care about is using plugins and transitions and being able to see them all play in real time, not all choppy and skipping frames. I haven’t seen one KZbinr explain what each new MacBook can and can’t do. They just say that the Max is faster. I care more about timeline performance when using proxies and whether or not I can perform certain functions with one machine over the other.

[@HeydoubleU]

Nice comparison, would be cool if the project files were available for download to try and compare results

[@philcretired5143]

Great comparisons. Thanks much. Downloadable project files would be terrific. Many people could do comparisons to their current machines.

[@joraphpictures]

Love the ideal to be able to download the project files..

[@LearnColorGrading]

That's a cool idea. I will try to do that, as the files are a bit large.

[@njasaa]

Laptop amd 5900hx/3080. Speed warp 4k timeline - 4k, H265, FPS 59.94 (Mavic). + grading + text + 20% slowed down. I'm getting 3-3.5 frames playback.

[@ririlan2137]

Since not one can playback in realtime, I assume that optimisation will be turned on and maybe rendering. So it would be interesting to see how long it takes to render a Speed Warp or a 'noise reduce' for a 4k clip of 2 minutes. It takes forever for my Macbook Pro 2018 to do those task. Wonder how fast it would be with one of those machine. (I love your course). :))

[@janriggert]

Yep, rendering time would be interesting. My weekly 4K BRAW news project (25 minutes long video with CC & NR + downscaled to HD) takes about 2 hours to export. Beeing able to export that in 30 minutes is extremly appealing!

[@petersndergaard5116]

I have M1 Max 14". Tried a 1 minute sequence with BRAW 4K, all clips with temporal NR 5 frames enhanced. Had other Open FX on my nodes as well. Export time was 2 minutes

[@denisl1071]

@@petersndergaard5116 What about fan noise on 14 max?

[@50shadesofNV]

Interesting series, unique to KZbin now, keep going.

[@tjpenton]

Respect for making these types of videos. It really helps the community. Cheers!

[@god-lightbulb]

In addition to just testing the M1, M1 Pro and M1 Max, it'd be useful to include a comparison of how a powerful PC would perform the same task.

[@nyambe]

Yes, it would be nice to know if these results are good, normal or poor. Also in normal workflow results with normal settings, at least to compare

[@keepout7766]

I would think the tests he was doing must be CPU intensive. If it was GPU intensive I would think the M1 Max would far outshine the regular M1

[@nyambe]

@@keepout7766 Sure, that is the thing the OG M1 with their 8 GPU cores are still pretty. Not in realtime but I have some amazing GPU intensive stuff on them. I mean stuff like Vray which they did pretty slow but they did it

[@Mionwang]

Exactly.

[@chfilm]

Or better so the Mac Pro!

[@italoamerighi3974]

finally...exactly what I was looking for. Have you done it on M2 chips as well?

[@gnkstudios6138]

This was very helpful thank you for putting in the time on these tests. Personally I would love to see if you can check the task manager (or whatever macOS performance checker is) to see what the machines are utilizing when under heavy load. It would help to know if there is a bottle neck happening. Other than that. Thanks again for putting these machines through the most difficult tests. Your tests certainly bring the viewer back to earth lol

[@alexanderzweifel9762]

Thank you for this! I just tested my 2020 Asus G14 (Ryzen 7, RTX 2060, 40GB RAM) along with these Macs. The G14 matches and outperforms the 16" MacBook in all of these tests. All this at a fraction of the price for the 16" MBP and with the ability to actually play some games. I am a PC user and I really don't agree with Apple on many things. The new Macs however, I have to say are quite impressive. But now I can rest easy knowing I already have an extremely well performing machine and the Mac wouldn't really be a quality of life improvement, especially considering the price!

[@theowlfromduolingo7982]

Can you do a video on the group function

[@xavier_arthur]

Thank you so much for doing this. Getting through all the hype & down to nuts and bolts it's very difficult to find

[@the.ansari]

Really love ur work and ur methods 👏🏼

[@clckristians]

Thank you for a very interesting test. I would be extremely interested to see how M1 / M1 Pro / M1 Max systems perform a longer (1-2 hour) multicamera (4-5) project? Appreciate that.

[@GlenReed]

👉Wonder how the system would handle ProRes with these same conditions, Especially the M1 Max since it has the two additional ProRes encode/decode engines. Thanks for the review.

[@VideoGameKillCounts]

I’m very interested in this as I use ProRes in my workflow a LOT!

[@JukkaLamsa]

External 4K HDR display and 4K HDR 60p timeline, does the machine clot?

[@gabydiaz4882]

Most of us work with optimized media, can you do the same tests but with optimized media? thanks for all your effort, Greetings from TJ!

[@MakeKasprzak]

Thank you for testing this. :D

[@Nicola0192]

This what i was need! Thanks

[@martagalanoficial]

Great video! You are testing what exactly matches my needs. I would love to see these tests using Neat Nideo noise remover and check how the Pro and Max are handling this plugin...

[@donghuilin7762]

Did you use the latest "5x faster' version of davinci resolve?

[@TruthReviews]

Great video thank you.

[@sethzhong]

It'll be really cool if there's a test for the high end M1 Max on Sony's 10-bit 422 XAVCS files with noise reduction - temporal only, and temporal and spatial together!

[@Teluric2]

No full test in any video..nobody uploads a full video test because the speed is not what benchmarks claim I dare anybody to post a video from start to end showing progres bar on screen.

[@sethzhong]

@@Teluric2 I own a M1 Max 16" now, can confirm it lags on Sony's 4k. Change to 1080p timeline and its smooth with nr

[@pacmin175]

Thank you for your time . I create home movies up to 1 hour long . Sold all cameras and now use iPhone 13 for video . Can the base model 14inch pro 2021 handle basic editing, grading, storage of 4K and pro res projects up to 1hour long, in other words can I finally ditch the desktop pc? Many thanks again. Pacmin

[@Theomachos]

I would really like to see a comparison of all 3 Apple devices vs. a "gaming" laptop, eg. the Razer Blade 15 advanced, preferably with some more complex fusion VFX.

[@77dris]

And I would like to see them both unplugged for the tests.

[@fnytnqsladcgqlefzcqxlzlcgj9220]

or a workstation laptop like a maxxed out precision 7740-50-60

[@Teluric2]

@@unknownuser069 maxtech??? Are you serious? He s a phony youtuber making reviews of m1max duo like he would have one.

[@MrMikePalma]

I would have loved to see same type of test, but with effect layers of more varying difficulty. But really nice tests here!

[@juanesilva]

I love your videos :3 very deep

[@Raidenorius]

Great test, but would be nite if you render these clips and show the time results

[@lowesmed5681]

On the 14 inch base model, if you could only upgrade one. Would you recommend going for 32GB RAM or the 10c/16GPU version upgrade? For Davinci work and using Adobe suite?

[@badGamr]

Awesome video! Test superscale next perchance?

[@OliverH93]

Great tests. Its videos like this ive been waiting to watch, thank you for the tests. I'd like to see how it handles a 30 minute 4k timeline with 3x h.264 video tracks, basic colour grade (with film grain) and audio edited too. I find my playback, scrubbing etc. on my macs slow down aa lot when I start to edit the audio with de-esser, multiband, voice noise reduction etc. A lot of the work i shoot is 30min-1 hour interviews with multiple angles, and my machines really slow down when its a 20 min+ project with all the additional things I want to do.

[@samuelgerkin]

Fusion testing would be awesome!

[@tablerockaerialphotography]

What CODEC is being used on the sample footage?

[@rahuldhadave]

Thank You So much for this useful Information

[@MusicForIMAGES]

Like you I use Resolve to produce videos. Do you think it is possible that the Mini M1 can write a Rendering of a 4k project containing CC and Optical Flow even if the same project does not start playback? Thank you for your time in understanding and solving our common problems.

[@JoshKoch]

Are you running the 16 inch Max on high power at all?

[@MrCjrodriguez]

Very interesting comparison seeing all the systems fail the stress tests. But you haven’t looked at the performance of the systems to encode and export the content. I’ve seen other demonstrations where there is significant time savings exporting the content. How weight would you give to the time saved for encoding and exporting of content?

[@WilliamTurnerViolins]

Thank you for such great and informative content. Also love your course, took advantage of a prior special and so impressed when it was updated for DaVinci Resolve 17. Serious respect!

[@helmoh]

Tnx for the comparisson. Could you also test differences when applying color correction or effects? And perhaps render times as well.

[@g0ral]

Could you share those test so we could compare our PCs to those macks?

[@mediatrix1111]

Sir, am buying a new Macbook, and am deciding between these 2 options.... 1). 15" M2 Macbook air [16Gb-512GB], or 2). 16" M1 Macbook Pro [16Gb-512Gb]. Currently the MBP is on sale and going for $1559 (512GB). Buying it for video (Davinci Resolve studio) and audio (Logic Pro). Which one should I buy? The MBP even has 1TB model going only for $1659/-. But since I plan to use an external SSD for editing, I have decided for the 512Gb model. Plz guide me. Thks.

[@eddietor_cine]

Very interesting. I understand about being torn. I wonder if on the software anything else can be done by Blackmagic with more testing to take more advantage of the new m1 and get better results ?

[@NowyKurs]

How this would preform on MacBook Air M1?

[@elkawash]

Thanks. But how do the new models perform against the best pc laptops?

[@1710000huh]

Can you please share this drt or drp? I wanna test it with 6k braw on 4k timeline on amd 6900xt

[@keliu5821]

Thanks a lot for the testing. I recently tried to do some tracking in 4K footage on my M1 iMac 16G version, it’s very slow, planer tracker is a bit faster but easily prompts me app ram shortage warning. Did you encounter similar problems? Native Davinci 17.4 free version.

[@fstarstudios24]

What laptop do you use??

[@simonwyndham]

For noise reduction that was impressive. My current machine can't even play at 1fps! I'd be thinking a few years down the line, because I'd want my machine to be able to deal with what's coming tomorrow rather than simply today, since I tend to keep my laptops for quite a long time before upgrading. From many of the tests I've seen elsewhere render times are a key area in which the new machines excel, with the M1 Max exporting noticeably quicker than the M1 and M1 Pro. Even a 1-minute reduction in render time is significant. Over 60 renders that's a whole hour that you could either be working on another edit, other work, or just closing up and finishing early. 60 renders isn't much over the course of a whole year, so actually over the course of that year you could literally be saving hours of time, particularly if your edits are medium to long length. If you edit a 30 minute video and it takes 30 minutes on one laptop and 15 minutes or faster on the newer one, imagine how quickly that time saving will add up over time. That's only 4 renders and you've saved an hour! Now add that up over the lifetime of the laptop, and the time saving could be extremely significant. And that's even more significant when you look at the noise reduction test. Because although you might be able to optimise things for realtime playback during editing, when it comes to exporting you will often want full quality. Not to mention, can you also do things at the same time? For example my current Intel MacBook will slow everything down whilst exporting from Resolve, so I can't get on with other things very easily. With the better performance of these new Macs I might be able to get on with other work.

[@n0f4ke74]

yea, noise reduction, i didn't see any noise reduction in his video .....

[@the.ansari]

Can you please do the same test but with medium optimize setting like in maybe a 1080p file or something ?

[@badboy350]

Awesome job Alex. Thank you. Is it possible to do a render test using noise reduction and speed wrap and and pit them against a rtx 3080. I know rtx 3080 is a beast.

[@nyambe]

Can you do the test with a high end windows laptop. what type of system can pass these tests?

[@senpera]

Would love to see the same test in FHD timeline. It should be a lot easier but workable easier or still chocking madly?

[@AndreGranjo]

Thanks for these videos!! Great content and very relevant. Have you tried to run the Puget Systems DaVinci Resolve' benchmarks? Since you have the 3 models this would be fantastic for cross platform comparisons.

[@jerrychiasson9857]

I’m a little confused is this program you’re using maximized for the M1 or is it running Rosetta?

[@SliwkaProductions]

I'm looking for Fusion test with 3D, particles etc. and how it performs in this type of animations :)

[@frassmicce]

Hello, is it also possible to run test using GPUs like RTX 3090?

[@nyneak]

pls, i need a test with h265 10 bit 422 from a7siii and export h265 fps resolve. thx

[@fredstork]

Very interesting, thanks for doing this! If you have the possibility to compare them running the magic mask it would be really interesting.

[@seancloutier2577]

Hey Alex, how does the magic mask work. Just curious.

[@actnreact1321]

dont get me worng you can most defenitely use render cache or optomization but for me i think it takes up WAY to much storage and you still have too wait for it to render out dependiong on how fast your machine is so for me atleast i would still go with the m1 pro which makes the most sense

[@EvolutioStudio]

Hi! How about other OFX? Grain, glow?

[@dorinxtg]

Thanks for the clip! I have a small question: I'm doing mostly vlogging, minimum color corrections, some titles, all in 1080p with clips are imported as H.264 or Prores LT. My question: Which M1 could be sufficient - the 2020 M1 8GB or 16GB?

[@MrBaptista]

Thanks for the insights. If you could do some tests editing and grading h.265 10 bit footage it would be great!

[@xavier_arthur]

Can you throw different codex in a timeline and stack some on top of each other, like some h.264 and h265 With no optimization. Maybe 10 to 12 minutes timeline

[@jraiber]

Thank you very much for this test. You saved me from wanting the new MBPs. I have the 2020 16" with the radeon 5600m build in and tested it with the same setting like you did with the newer ones and guess what, it's faster. Yes, for example the toughest spatial NR is played back with my machine with 10 fps. (everything the same, 4k, no optimisation etc). To push things further, I normally use an egpu for grading. ( a sonnet breakaway box with a 5700XT). The egpu alone make 14fps (twice the speed of the new high end-model) and internal+external gpu make 19 fps. compared to your reached 8fps. Isn't this interesting? If I want to improve my 4k workflow I maybe go with another egpu maybe a big navi one. This are good news for me, because of all the KZbinrs, I thought ok, my machine is completely outdated after a year. But it's not. Not for Davinci, which is my most performance hungry workload. Thank you for this Information.

[@Theaaronchris]

Hey, I’m looking to get an egpu to handle NR for some of my work. Does the egpu only help with playback? I’m trying to figure out how to get the fastest export render times. Does the egpu help speed up final render times as well?

[@Theaaronchris]

Thanks for your comment.i have a 2018 MBP with Radeon 555x with 4GB of vram and I was wondering if adding an egpu would help and from your comment it will!

[@Jonasdelrey]

Thanks, I’m going to buy the cheap 13 inch!

[@77dris]

It's weird all the differences I'm seeing in benchmarks on KZbin. Some people get great results (almost game changing), others are getting "meh" results at best. I watched an iJustine video a couple days ago and she had a video guy come over with 8K raw footage and he was applying luts and noise reduction and couldn't get it to hiccup at all. I've seen people game and get better results on their M1 Max than on their RTX 3080 laptop. Others are getting mediocre framerates. And it's consistent for each channel. If one channel gets great results, it always gets great results. If a channel gets poor results, it's always getting poor results.

[@TheCHEATER900]

I'm just gonna assume iJustine used final cut pro rather than resolve and 8k prores, right? Because it's specifically made for that.

[@jujuakajuliuskivi]

@@TheCHEATER900 they used resolve

[@branpod]

They likely used the optimizations. This video is specifically intended to bring the macs to their knees. Realistically you'd never actually be working like this.

[@EarthKingPrime]

Hello, i want to know if it can handle heavy 3D fusion comps smoothly? Fusion usually is not for major 3D comps so how far can you take them? Apple boasts 5x speed so i want to see smooth playback of it

[@NateDulcimer]

How would you compare the screens of this year's models to last year's? Does the Micro LED technology make the screens better for color grading? I don't expect that they are able to be calibrated, but maybe they are better out of the box?

[@GQLewis]

Was the M1 Max in high performance mode?

[@ledbypassion]

Thanks a lot for these tests. I must admit I’m quite disappointed with the performance. Let’s be honest I didn’t expect any of them will be able to play Speed Warp in real time but the playback performance gives a good estimation of how long the smart optimisation will take and it’s approximately 25 times the duration of your speed warp clip

[@boodi1170]

I'd love to see what type of system that can play speed warp and noise reduction in real time I have a 3090 and it plays speed warp but its choppy

[@Teluric2]

@@unknownuser069 multiple 3080? Evidence?

[@hellowallace]

With these tests, which hardware could playback smoothly live? Mac Pro with multiple GPUs and After burner? or nothing at all?

[@aviationodyssey1892]

It would be great to see how fast they render cache for such a clips to get the real time playback (with no other optimization). Also I would be grateful for h.265 4K or 6K source video samples

[@Nattodayy]

I haven’t seen very many After Effects tests if you’d be willing to do some of those.

[@melgross]

So far, all of,the test done here with these new models have been just about the hardest tests Resolve can manage. I’d like to see these in comparison to an equivalent Windows machine-all running on battery. Otherwise, these tests are being done in isolation with no context. Are there any laptops that can perform these test while on battery (or even plugged in?

[@HunterOfMyFate]

I bought the M1 Mac mini based on how powerful everyone said it was. My kids love making videos in front of the green screen and then I'm left to edit them. I have the 16GB version, I edit at 1080p 60fps because they jump around so much. I use smart render cache and proxy but green screen and special effects in fusion is still so slow. My question isn't do I need the M1 Max. My question is if the M1 max is powerful enough for what I need. Could you please test these systems with green screen and fusion special effects? From a realistic user experience, with smart render cache, would the 400MB/s memory bus on the pro max feed the cache noticeably faster during editing? I've read that if you step up to the 2TB or larger SSD you get a significantly faster SSD with 7500 GB/s read and write speeds vs 5200 GB/s on the 1TB and smaller. Would that have a noticeable difference in a typical workflow in Resolve?

[@mkanoute]

I fail to understand the point of performing tests in non realistic scenarios (like with no optimization). I understand the waiting for proxies to be generated can slow down by a lot the workflow in addition to consuming disk space, so checking whether one can do without is valid. But for the rest? You could at least check realistic options like rending in the background, editing at a lower resolution, lower quality, etc.

[@Akkigraphy]

Thiis is actually a very good testing. Quite practical scenarios that your would use or come accross while wroking on a project. If one is not interested in Mac, which laptop would you recomend for 4K editing projects?

[@ademmedia]

Test it in the fusion in hard scenes, pleeeeease ))

[@fanjan7527]

The M1 Air is a big upgrade for me, when I add any noise reduction on my 3 year old PC I get about a 9s lag between frames.

[@janislionel1715]

Would be really interesting to see the following as a professional: (1) How would 32gb compare to you 64gb of memory on the Max? (2) How would the tests perform on the Max with one or two monitors attached?

[@Stargaze79]

I have an Asus ROG Strix with a Ryzen 9 5900HX and RTX 3060. I wonder how it stacks up to the M1 Max.

[@drewarmstrong4325]

This is fantastic. BUT… to really figure out what the differences are you would have to level them based on screen size, sound system, built in cameras, speed of memory etc and yes, if you were strictly buying a machine to edit video in resolve on an exterior monitor. You got it. Hard to justify the extra money. However, for 90% of us out here in the real world the decision is not distilled down to that. Now, if I was buying my college student a computer for school that needed to be able to run photoshop and resolve… OK, I would for sure get the straight M1. But I am a guy who has high end cameras and drones and love using those. I shoot in 4K or 5.7K or 8K when I could REALLY be just fine shooting in HD. Why do I do it? Because I want to. Does it tax my current hardware in terms of processing power and storage for doing so? Yep. But I am good with that. And, frankly, I can afford it and this is my thing while high end bikes or sports cars are some other persons thing. But you did say this. I am just expounding.

[@sharonb.9128]

This test initially seemed valid, comparing Apples to Apples, BUT because they all did so poorly the obvious question now is “compared to what”? You REALLY need to show the laptops that can pass this test. If none can, then you have provided an unrealistic benchmark in a vacuum for Apple laptops.

[@BenMilford]

Yeah maybe tests to compare generating optimized media. Or something to show how long it would take to get things to be playing back in real time.

[@PeteManabat]

Finally. 👌

[@mclarenf1V12]

Yeah, try and optimize all your media on the M1 compared to the $5000 M1 MAX. I know my Max optimizes in seconds. So.......... there is more to consider than just play back. Also, what computer can play Speed Warp in real time? that would be a good baseline to know.

[@baredesigns1]

However M1 ultra fails to be as fast in video editing, rendering and export when compared to an i9 with RTX 3080 Card 64GB setup while working on an 4K and 8K video

[@denisl1071]

No sense to buy any of them, PC Puget still is better?

[@christiangillmann9466]

These machines "fail" to playback this very unreal workflow scenario in real time compared to what machine? I did not get your point. If you work professionally with colour correction you want a faster machine, maybe even a mobile fast machine. So do these modals dissapoint in doing that? I guess not.

[@Ttwgkn]

Render on Monterey and Big Sur , please

[@OnlyVocals]

Why? Is Big Sur faster?

[@Vortex124875]

..if you had the 14" 2021 macbook pro model equipped with M1Max it would outperform the 16" M1Max model due to its smaller screen (it doesn't waste GPU resources/memory to drive the larger screen) so you can get even better performance from a Macbook pro than this video shows.. edit: sorry my assumption was wrong see comment below..

[@denisl1071]

It is very noisy

[@branpod]

The 14" 32 core is actually throttles and performs worse than the 16" 24 core

[@Vortex124875]

@@branpod ah good to know

[@Siimkristjanpariis]

Heres an idea: Share us the link of the exact footage you used in these test. Im curious to see how my old desktop PC will react with the same effects applied.

[@laurenthenry]

Please add timestamp so we can skip to what we need 🙂👍

[@ericemerick5065]

These tests conclude that you should not be editing h.26x files in a professional world, which most of us learned years ago. It’s a delivery codec, NOT A CAPTURE/ORIGINATING CODEC!

[@imiy]

why to even edit in 4k timeline? that's just an unrealistic experiment. what's the point to compare those models between each other? it's obvious which one is more powerful. better to have a real understanding on how they perform in real world tasks. so one could understand which model is enough for him/her.

[@Clickumentary]

Ok I'm convinced there is *no* silver bullet with M1 chip. Yes, better, but not worth spending the cash on for a portable solution. I'm sticking to wintel and setting sights on Puget for desktop.

[@denisl1071]

Probably it is right.

[@RolandsBriedis]

Stabilisation.

[@innocent81man]

Dear sir. If NO ONE would edit a video in those settings so it’s so logical that the manufacturers won’t design their products to work in such conditions! So I honestly don’t understand the point of your videos?! It would be really useful if you would test those products in real life scenarios and in the settings that are useful and commonly used so we as buyers can see the improvements those Apple products have gained and if there’re worth the extra money Apple are asking for! So much thanks for your effort.

[@ledbypassion]

These are real scenario, even if you don’t want them to play this in real time, once you use one of these effects in one of your editing, even if you use optimisation, the first time of playback is often in ‘real time’ which do the caching at the same time. If you need 2mn for a 4s clip duration for speed warp it’s really really annoying. In addition, this scenario, on my M1 MacBook Pro just make it freeze during the whole process, so it’s even difficult to change the setting of the optical flow

[@innocent81man]

@@ledbypassion I was talking about the optimizations of which he himself said that ALL of the three laptops would work just fine when applied of which is the way people usually edit their videos!! So why testing the laptops in a way that NO ONE would use! Like, the technology is there to be used and people go further to the next level from that point, right! It’s like checking the time it would take to walk between to cities when there’s an invention called “Car”! No one would make a use of the outcome results cause no one would walk that distance, right! Thanks anyway 👍🏼