HomeVideos

MacBook Pro M5 vs M5 Pro - These Results are INSANE!

Now Playing

MacBook Pro M5 vs M5 Pro - These Results are INSANE!

Transcript

287 segments

0:00

So, then, guys, a lot of you are

0:01

wondering, how much more powerful is the

0:04

baseline M5 Pro compared to the baseline

0:08

M5 MacBook Pro? Well, today, what I want

0:12

to do is I want to compare both of these

0:14

models. I've got the likes of an M5

0:17

MacBook Pro here, and I've also got the

0:19

M5 Pro MacBook Pro with the baseline

0:22

configurations. This is the one with the

0:24

lowest amount of CPU GPU cores, and also

0:27

the one with 24 GB of RAM, 1 TB hard

0:30

drive. And then the same here, this is

0:32

the same standard M5 chip, the one with

0:34

the 10-core CPU, 10-core GPU, also 1 TB

0:38

storage, but 16 GB of RAM. So, today,

0:41

what I'm going to do is I'm going to

0:42

compare both of these machines to see

0:45

what they are like in performance-wise,

0:47

cuz design-wise and everything, they're

0:49

pretty much identical. So, let's take a

0:52

look, then. And the first thing I

0:53

decided to do for you guys is I decided

0:55

to compare the storage speeds. Let's see

0:57

what the difference is that we have

0:58

right here. Well, we can see that

1:01

straight away, using Blackmagic read

1:02

speed, that obviously the likes of the

1:04

M5 Pro is faster. And this was expected.

1:08

At the end of the day, it's using PCIe 5

1:10

here for the M5 Pro for the storage

1:13

speed, so it's far, far faster, and you

1:16

can see the difference here. We are

1:18

talking around about a 91% increase in

1:22

just read speed, and about an 81%

1:24

increase in write speed on the M5 Pro.

1:27

Now, what I am going to shout out here

1:28

is the M5 standard also got a massive

1:31

boost this time round when we got it

1:34

released back in October last year. It

1:37

used to be around again about half this.

1:39

It used to around to be about the 3,000

1:41

kind of mark there, 3,000 MB. So, even

1:44

at 6,000 MB, so that is super fast.

1:46

Remember, the M4 Pro is around about

1:48

that speed, and that's no slouch,

1:50

either, with its read or write speed.

1:52

But, just to let you guys know, just the

1:54

difference here that obviously the M5

1:55

Pro is definitely way faster, but

1:58

whether you're going to utilize all that

2:00

speed, that is the big question here. I

2:02

don't think many people are going to

2:04

utilize that fast disk speed, especially

2:06

with the chip that's inside of this

2:08

machine. But then, after this, the next

2:10

question is, what about CPU performance

2:13

in both these machines? I think this is

2:14

one of the big key differences.

2:15

[clears throat] So, obviously, you got

2:16

the M5 Pro and the M5, and I ran things

2:19

like Geekbench. I've also ran things

2:20

like Cinebench to see what we got in

2:22

single and multi-core score. Have a look

2:24

at the results right here. So, you can

2:27

see that obviously in single-core score,

2:29

they are basically identical cuz at the

2:30

end of the day, they're using the same,

2:32

you know, single cores here. They You

2:34

just get more of them in the M5 Pro. So,

2:37

yeah, essentially, it is literally neck

2:39

and neck here. Nothing really whatsoever

2:41

any differences. But in Geekbench

2:43

multi-core, obviously, you can see the

2:45

difference here. That obviously, we're

2:46

getting about a 56% increase here in

2:49

speed in multi-core performance on the

2:52

M5 Pro to the M5. But do remember again,

2:55

what we're getting here on the M5

2:57

standard, this was very similar to what

2:59

we got with the M4 Pro, you know, this

3:01

time last year. So, yeah, it's still

3:03

super fast. And you know, it's amazing

3:06

to think that even though when we had

3:07

the original M1, this used to get like a

3:09

score of about 8,000 or 8,500. It's

3:12

about the same speed now that we get

3:14

with say a MacBook Neo. So, it's crazy

3:17

to think how much has increased over

3:18

those 5 years. But yeah, what I would be

3:21

saying is that obviously, there is

3:22

definitely an increase there. It's the

3:23

same with Cinebench as well. Exactly the

3:25

same score for single-core, you know,

3:27

727.

3:29

But for the multi-core score, that

3:31

obviously, yeah, you are seeing an

3:32

increase there here up to 61%.

3:35

Obviously, Cinebench and Geekbench run

3:37

their benchmarking tools differently,

3:39

but yeah, this is what we're getting in

3:41

the speeds. But then, the next question

3:43

is, what about say GPU performance?

3:45

Well, again, I've done the likes of

3:46

Geekbench. I've also done Steel Nomad

3:48

Light, and I've also ran Cyberpunk 2077

3:52

on both of these machines at high

3:53

settings with their metal effects for

3:55

built for the machines. And let's have a

3:57

look to see what we got that in scores

3:59

and frames per second.

4:01

So, we can see that with Geekbench

4:02

metal, that obviously the M5 Pro, yeah,

4:04

it's always going to be out in front

4:06

here. I'm going to say it. So, no

4:07

surprises, but how much? It's about 84%

4:10

more quicker than what we had before.

4:12

And the same with still Nomad light in

4:14

the matter of frames per second. This

4:15

actually uses quite a lot of that ray

4:17

tracing kind of cores and bits and

4:18

pieces. But, yeah, we got 66.6 frames

4:21

per second compared to 38.5.

4:24

That is a big boost what we are getting

4:27

here. Mainly because we have so many

4:29

more GPU cores inside the likes of the

4:32

M5 Pro. But, there again, with the

4:35

Cyberpunk 2077, obviously there is a

4:38

massive boost here as well. It's over

4:39

double the amount of frames we are

4:41

getting here. 110 frames per second

4:43

compared to 54. Again, I think it's all

4:46

of those extra sort of GPU cores. What

4:49

we are getting here it is making a big

4:51

advantage for this machine. But, then

4:54

next of all, you guys are probably

4:55

wanting, "What about some AI bits and

4:57

pieces? What's happening there?" Well, I

4:59

ran things like Diffusion B, as you can

5:01

see here to generate a picture of

5:03

English countryside on a sunrise. And I

5:06

also ran, you know, the latest Llama 3.2

5:09

free beer to see how many tokens per

5:10

second we got. And here are the results.

5:13

We can see that in Diffusion B, that

5:14

obviously, you know, quicker here. So,

5:16

slower, the smaller the chart is better

5:18

in this sense. So, we can see the M5

5:21

took 12. You know, 6 seconds to create

5:24

this picture. It's not bad at all. But,

5:26

the M5 Pro took 7.3 seconds to generate

5:30

the exact same picture. That's a 42%

5:32

increase. Whereas where we got with

5:34

Llama, then obviously, you can see that

5:36

obviously we are making more tokens per

5:38

second. The M5 Pro, it's about a 45%

5:41

increase there. So, you can see it's

5:42

around about the same amount of

5:43

percentage increase for AI capabilities

5:46

on both of these machines. But then the

5:48

next question is what about creator

5:50

workflows and creator workspace? What

5:53

about things like Blender with Party Tug

5:56

and also the likes of say let's say

5:57

exporting 10 raw pictures into JPEG and

6:00

also running things like Final Cut Pro

6:04

going into ProRes 422 and also into

6:07

HEVC. Well, I've run these tests and

6:09

let's have a look at the results right

6:10

here. Well, we can see straight away

6:12

Party Tug it took 57 seconds to render

6:15

this image whereas M5 Pro took 37

6:18

seconds. So yeah, 20 second difference

6:20

there is far faster. Same with

6:22

Pixelmator obviously the 10 raw pictures

6:24

to JPEG it took 88 seconds on the M5 and

6:28

the M5 Pro took 60 seconds. That's more

6:30

CPU bound and then obviously the kind of

6:33

export that we did in Final Cut Pro 422

6:36

8-minute video export M5 did it in 39

6:39

seconds M5 Pro did it in 26 seconds. So

6:43

yeah, that is definitely faster there.

6:45

But then look at the HEVC export there.

6:47

This one's not so different because they

6:48

use basically the same media engine

6:51

cores. This is more using the memory

6:52

bandwidth to the advantage I'd say more

6:55

cuz obviously the media cores are

6:56

exactly the same in both these machines

6:58

but you can see that obviously the M5

7:01

actually managed to export this in 194

7:05

seconds and then the M5 Pro took 208

7:09

seconds. So what was quite interestingly

7:11

was that the M5 Pro was a little bit

7:13

slower. Now just in case you want to

7:14

know I ran this test three times and

7:17

this was the median score. But whether

7:19

this is a glitch or something I'm not

7:21

too sure but what I'm basically saying

7:23

this is the closest result. Really

7:25

there's not going to be much difference

7:26

here in the HEVC export. That's what I'm

7:28

just going to say there. But then after

7:31

doing all these tests you guys probably

7:32

want to know well how does thermals you

7:34

know do in these machines cuz obviously

7:36

the likes of the M5 MacBook Pro only has

7:39

one fan inside whereas the M5 Pro well

7:41

we actually have two fans inside. So,

7:43

let's see what we got with cooling. So,

7:45

what I did was I ran with Cinebench,

7:47

again doing a 10 minute, and as you can

7:50

see right here, I checked out the

7:51

internal sensors, and I also checked out

7:54

the likes of say doing a thermal camera.

7:57

And let's look at the results. So, what

7:59

we can see right here that obviously we

8:01

push this all up that the M5 was about

8:04

105°, but so was the M5 Pro on the

8:07

internal sensors. This is what it was

8:09

saying, but when I was using the thermal

8:11

camera to actually measure the middle of

8:13

the keyboard area, you could see the M5

8:16

was actually ever so slightly hotter, 1°

8:19

C

8:19

more or 2° F

8:21

more if that. And yeah, it was nothing

8:24

really in it whatsoever. Obviously, the

8:26

M5 Pro is more powerful, generating more

8:28

heat, but you do have those two fans.

8:30

Whereas, the M5 doesn't create as much

8:32

heat cuz it's not as powerful, but

8:34

there's only one fan. Really,

8:36

personally, I would say it's neck and

8:38

neck here. But, the next question you

8:40

want to know is, well, how well did

8:41

battery life do after all of these

8:43

tests? When I first started these tests,

8:45

both these machines had 100% battery.

8:47

Also, the battery counts on these were

8:49

very similar to each other. So, yeah,

8:51

not much difference there. But, let's

8:53

have a look at them the percentage after

8:54

running these tests what we had.

8:57

Well, we can see that the M5 had 43%

9:00

battery, whereas the M5 Pro, well, that

9:03

had 35%

9:05

battery left. So, definitely the M5 has

9:08

a better battery life cuz obviously it's

9:10

only got one fan. Also, it's less cores

9:12

as well. So, yeah, it's definitely

9:15

better in that sense. But, what I would

9:18

say to take away from all of these

9:19

testings today is that obviously, yeah,

9:21

paying that $500 more does make a

9:24

difference, but it's not saying that the

9:26

M5 is not capable. The M5 is more than

9:28

capable of doing a lot of these things.

9:30

Look at that Diffusion B example that I

9:32

was talking about earlier. You know, we

9:34

were talking just a matter of seconds

9:36

difference. It's the same with a HEVC

9:37

export and also, you know, the ProRes

9:40

kind of export there. There was not much

9:43

in it whatsoever. It really, really

9:45

depends, you know, what you're going to

9:46

be using these machines for. If you are

9:47

going to be all day constantly using pro

9:51

apps non-stop, then obviously the M5 Pro

9:54

is going to be good. But if you're

9:55

using, say, pro apps a couple of times a

9:57

day or, you know, once or twice a week,

10:00

then obviously, you know, even the M5 is

10:03

going to be absolutely fine. Or even

10:04

MacBook Air, you never know, that might

10:06

even be fine for you, too. But the main

10:08

difference is that you have seen what we

10:09

are getting here in the M5 Pro and the

10:12

M5. But really, guys, that is it. It's

10:14

time to wrap up the video. So, if you

10:16

have enjoyed watching it today, please

10:17

do press the like button. As always, you

10:19

want to hear the latest Apple news,

10:20

reviews, and comparisons, make sure you

10:22

subscribe to the channel and hit that

10:23

notification bell. Until next time,

10:25

guys, I'll see you really soon. Take

10:27

care. Bye-bye.

Interactive Summary

This video provides a comprehensive performance comparison between the baseline M5 MacBook Pro and the baseline M5 Pro MacBook Pro. Through various benchmarks including storage speed, CPU/GPU tasks, AI generation, and creator workflows, the host examines whether the higher-tier model's performance justifies the extra cost, ultimately suggesting that while the M5 Pro is faster, the baseline M5 is highly capable for most users.

Suggested questions

3 ready-made prompts