What's wrong with my Mac?
Why is it lagging like this? Hello. This is Seoulian. The editing laptop I'm currently
using as my main device is the MacBook Pro 16-inch M1 Max. Now, the MacBook Pro
M2 Max is also available, and recently, high-performance
laptops with RTX 4090 for Windows have been released as well. So, I decided to look into what I should
replace my main editing laptop with. I'll bring two laptops, each
delivering top-tier performance in the macOS and Windows worlds, run benchmarks, and conduct
some tests in Lightroom and DaVinci Resolve that I
currently use, to compare and see which
product is more suitable. First, in the macOS camp, we have the
14-inch MacBook Pro. I wanted to test it before
purchasing, so I borrowed it from a
knowledgeable friend. It's a model upgraded
with an M2 Max 12-core CPU and a 38-core GPU
processor. It has 96GB of unified memory and
4TB of storage space. Except for the storage space,
both the processor and unified memory have been
upgraded to the maximum, so it should be the top-
spec in the macOS laptop world, right? The price is 6,990,000 KRW. In the Windows camp,
MSI has entered the arena. It has an Intel 13th-gen
Raptor Lake i9-13980HX processor and a GeForce RTX 4090. It has 32GB of memory and
2TB of storage space. The price is 5,400,000 KRW. Even if the specifications
are similar, laptops can have performance differences depending on
available power. This product has a
maximum of 250W, so considering
the available power, if it there is a high possibility that it will outperform other laptops
with the same specs. Alright, no need for long
speeches, right? Let's start with a quick benchmark, and then move on to
real-world tests in Lightroom Classic and DaVinci Resolve,
tailored to my usage. First, I ran the easiest Geekbench
for comparing different platforms. Both laptops are currently powered on. The Windows laptop is set
to maximum performance, and it's configured to use only the RTX 4090 with the MUX switch, regardless of the battery. Looking at the CPU test results, the MacBook Pro scored
2,697 points for single-core and 14,888 points for multi-core, while the Windows laptop scored
2,663 points for single-core and 13,847 points for multi-core. You can see that the MacBook Pro slightly
outperforms in both single and multi-core tests, with about a 6% performance
difference in multi-core. On the other hand, in the GPU test, the
M2 Max scored 139,209 points, while the RTX 4090 scored
204,266 points, showing a whopping
46% higher score for the RTX 4090. It's truly impressive, isn't it? The performance of the RTX
4090 is remarkable, and this laptop, which can deliver this performance
with a TGP of 250W, is quite impressive as well. Next, I ran Cinebench, which puts a
heavy load on the CPU. Although it's a test where
Intel scores well, I was curious to see what score it would
achieve under high CPU load for 10 minutes. The M2 Max scored 14,039 points, while the i9-13980HX scored 27,261 points, showing a significant lead for Intel
with nearly double the score. Cinebench is a test that
allows us to see how clock speed and power consumption stabilize and
remain in certain ranges during prolonged CPU load, similar to real-world usage for a little longer
than Geekbench. One interesting observation was that during the 10-minute test, the M2 Max
consumed a maximum of 40W and averaged around 35W
of power consumption, while the i9-13980HX
consumed a maximum of 186W, averaging around 115W
of power consumption. It's using more than
three times the power, but when you compare scores per
average power consumption, the M2 Max scored 401 points, while
the i9-13980HX scored 237 points, clearly emphasizing the strength of Apple
silicon in terms of performance efficiency. People may not buy a
MacBook Pro for gaming, but I still ran 3DMark Wild
Life to assess the graphics card gaming performance
across platforms. It allows for comparisons between
Windows, Mac, and iPad, making it useful not only for gaming but also for comparing performance
differences across platforms. The M2 Max scored 90,811 points, while the RTX 4090 scored 119,906 points, showing a 32% higher score for the
RTX 4090 compared to the M2 Max. It continues to follow a similar pattern, with the M2 Max having
an upgraded GPU with 38 cores and ample unified memory, but the laptop-grade RTX 4090 truly shines in terms of
graphics performance. Next, I measured the perceived
speed in real-world usage. Recently, in Lightroom Classic, I've been using the AI Noise
Reduction feature for almost all of my photos because the
results are very natural and pleasing, but it puts a significant
load on the system. I measured how long it takes to apply
AI Noise Reduction continuously to 19 photos
captured with the Hasselblad 907X 50C,
each with 50 million pixels. The MacBook Pro took 6
minutes and 53 seconds, while the Windows laptop only
took 2 minutes and 55 seconds. I was truly amazed at how much
faster the Windows laptop was. In Lightroom Classic, hardware
acceleration with RTX 4090 is possible, and I didn't know it was
optimized this well. It completed in less than half the time,
and I was really, really surprised. This was one of the most frustrating
processes when using Lightroom Classic on Mac, and now it's cut down by more than half? That really got me fascinated. It indeed did. Next, I measured the export time
for JPG at maximum quality. The MacBook Pro took 14.79 seconds, while the Windows laptop took
11.37 seconds, once again, slightly faster. The hardware acceleration with RTX 4000
series in Windows seems to be really good. I wonder how it would perform with
desktop-grade graphics cards. This also suddenly
piqued my curiosity. Next, I tested DaVinci
Resolve, the video editing software I
currently use as my main. If I had to choose the most frustrating
process in DaVinci Resolve, like in Lightroom, it would be the Denoise feature. After applying it with
high-quality settings, playback becomes slow
and really frustrating. I checked the playback frames for both with
hardware acceleration enabled. The MacBook Pro managed 22-24 frames, while the Windows laptop
achieved about 30-34 frames, clearly providing a smoother
playback experience. I'm testing it with a 60fps
4K footage, and as long as it consistently plays above 30fps, I can comfortably continue editing. Finally, I checked the
H.264 rendering time. The M2 Max took 49.55 seconds, while the Windows
laptop was slightly faster at 43.21 seconds. In terms of rendering frames per
second, the MacBook Pro was around 90fps, while the Windows laptop
achieved around 120fps. Wow... if even DaVinci
Resolve performs like this, shouldn't the next laptop just be a
Windows one? .. Comparing a 7 million KRW MacBook Pro with a 5.4
million KRW Windows laptop, and the Windows laptop is
performing like this? No way! I should go with Windows! If you were having similar thoughts, hold on a moment! There's one test we can't forget, right? Since it's a laptop,
people buy it to edit videos even on the KTX
train or on a plane, right? This heavy adapter is needed to
deliver performance. " I can carry this no matter what!" ...
even if you think that, you need to check how it performs in an environment where it uses
the battery, not power supply. The tests were conducted
with all batteries at 80% or more to ensure maximum performance. I'll show you just the Windows laptop battery
usage scenario scores from the results we tested earlier, because the MacBook Pro
had all the scores remain the same even
during battery usage. Geekbench 6 CPU multi-score
dropped by more than half to 6,751 points. Geekbench 6 GPU score was 52,191 points, almost a quarter of the previous score. And Cinebench scored 9,205 points, which
is about a third of the previous score. By the way, the Cinebench
10-minute test used 16% of the battery. I was really surprised that
the score dropped by a third even with this
much battery usage. 3DMark Wide Life scored 5,129 points. Wait, is it possible for it to drop by
almost 20 times? I was suspicious, so I ran it multiple
times, and it stayed the same. Since it's a multi-platform test,
it may not be accurate. Please take this into consideration. In Lightroom Classic AI Noise, it took
about three times longer, taking 8 minutes and 46 seconds. Rendering speed also increased by
more than twice, taking 24.08 seconds. In the situation with DaVinci
Resolve's Denoise effect, the playback frame rate dropped to
around 12 frames per second, almost half, and the rendering speed was 88
seconds, exactly twice as long. Today's conclusion is clear. The performance you can get from the ultimate
laptops in the Mac and Windows worlds is slightly better on the Windows side. However, if you plan to use the laptop for its
main purpose, photo and video editing, while on the go, relying on the battery, then the MacBook Pro is the only choice. The performance difference between power
supply and battery usage in Windows laptops is so significant that it's challenging to consider
them as the same hardware. In fact, on one hand, for a laptop with a
maximum available power of 250W, it probably can't maintain that performance
on battery for even 30 minutes. However, if the reason for purchasing
the ultimate Windows laptop is to have a single main system while
working from home and the office, using power supplies in
both locations, with all the peripherals like monitors,
desktops, keyboards, and mice all in one place, then buying a Windows laptop instead of a
MacBook Pro could be suitable for your needs. (Sigh...) Suddenly, after testing, I've decided to put aside the idea of
switching to a Windows laptop for now. Before returning the M2 Max to my
friend who loaned it to me, I want to do a real-world
comparison with my M1 Max and contemplate whether to
keep it or not a bit longer. Please leave any questions you have, and I'll greet you in the next video. Goodbye! End! Well, they're both beasts, aren't they? The world has really improved a lot. Right? But should I control myself this time? Why do I keep hesitating
every time, I wonder?
