15" Macbook Pro (2015) vs 15" Macbook Pro (2019)

[Tech_Mac_Man]

Hello Gents/Ladies,

Can someone who has both of these machines MAXED out with SPECS tell me what is going on here?

How in the world is the 2015 2.8ghz 16GB renders a video faster than a 2019 2.3ghz 32GB ???

Am I missing something here? I mean the 2015 has the same processor as the ones in the 2013 Macbook Pros which makes it 6 years old!!...Is the 2019 really worth buying?...If the 2019 is just a tiny bit faster than the 2015 in the real world??...What is going on here here exactly...??

Please see the video below so you know exactly what I am talking about...

 

[Neilfau]

The 2019 will definitely be faster than the 2015, look at the Geekbench scores. I think the issue with the test in the video is he’s using an external SSD to run the tests, but there would be a bottleneck from the usb 3. If he run the same tests from the internal SSD on each machine I think you would see a bigger difference.

 

[theapplehead]

You also have to consider the heat throttling going on with the processors of the 15” 2019 i9. That is a significant reason why the 2019 would have a slower processing speed than the 2015. Apparently the only time a user of the 2019 was to achieve the i9 turbo boost was to place his MBP in a freezer to keep it from overheating and throttling back the processor. Still not right of Apple to do this by any means. However, this seems to be the explanation.

 

[OSX15]

You also have to consider the heat throttling going on with the processors of the 15” 2019 i9. That is a significant reason why the 2019 would have a slower processing speed than the 2015. Apparently the only time a user of the 2019 was to achieve the i9 turbo boost was to place his MBP in a freezer to keep it from overheating and throttling back the processor. Still not right of Apple to do this by any means. However, this seems to be the explanation.

Is it still overheating? The Youtube videos on the 15 inch 2019 model seem to suggest much better performance than 2018 and cooler temps. Am I missing something here?

 

[pshufd]

The software could be better optimized for the older processor.

 

[theapplehead]

Is it still overheating? The Youtube videos on the 15 inch 2019 model seem to suggest much better performance than 2018 and cooler temps. Am I missing something here?

I believe I’ve still heard some lingering reports of throttling. However, I’d agree it is quite less than what was seen in previous iterations of the 15” MBP

 

[Tech_Mac_Man]

The 2019 will definitely be faster than the 2015, look at the Geekbench scores. I think the issue with the test in the video is he’s using an external SSD to run the tests, but there would be a bottleneck from the usb 3. If he run the same tests from the internal SSD on each machine I think you would see a bigger difference.

I was thinking the same...Maybe it had something to do with that external drive. But why would it be slow on the 2019 Macbook pro? I mean shouldn't it be faster as it is using a USB-C? It should be a 10x times faster transfer time than the USB 3.0? Just doesn't make sense here. We are talking about a 6year laptop. What is Apple trying to do? Go backwards in technology..?
[doublepost=1563151829][/doublepost]

You also have to consider the heat throttling going on with the processors of the 15” 2019 i9. That is a significant reason why the 2019 would have a slower processing speed than the 2015. Apparently the only time a user of the 2019 was to achieve the i9 turbo boost was to place his MBP in a freezer to keep it from overheating and throttling back the processor. Still not right of Apple to do this by any means. However, this seems to be the explanation.

There's heat issues with the NEW 2019 Macbook Pro's? I have never heard of it until now...They still haven't figured that simple solution out yet? It doesn't take 4-5 years in Engineering to under stand how to cool a laptop...C'mon Apple these are things a 10yr old knows how to do when building their own computer...COOLING is everything...!!

 

[ET3SW]

I was thinking the same...Maybe it had something to do with that external drive. But why would it be slow on the 2019 Macbook pro? I mean shouldn't it be faster as it is using a USB-C? It should be a 10x times faster transfer time than the USB 3.0? Just doesn't make sense here. We are talking about a 6year laptop. What is Apple trying to do? Go backwards in technology..?

That’s not how it works. The throughput of the drive is what matters. Just because the port is much faster does not mean the throughput of the drive is going to be any faster. If the drive can only read/write 500MB/s for example, no matter how fast the port is, it will max out at 500MB/s

 

[pshufd]

That’s not how it works. The throughput of the drive is what matters. Just because the port is much faster does not mean the throughput of the drive is going to be any faster. If the drive can only read/write 500MB/s for example, no matter how fast the port is, it will max out at 500MB/s

In general, computing runs at the speed of the slowest thing in the process

 

[iPhysicist]

Can I chime in and talk about things I don't understand? Because this is what this YouTuber does. Its the SSD that throttles the system. you can't export faster than it is able to write the flash storage. He could have used the activity monitor app and would have seen the i9 idling.

 

[Ploki]

The 2019 will definitely be faster than the 2015, look at the Geekbench scores. I think the issue with the test in the video is he’s using an external SSD to run the tests, but there would be a bottleneck from the usb 3. If he run the same tests from the internal SSD on each machine I think you would see a bigger difference.

In general, computing runs at the speed of the slowest thing in the process

well no. The disk is only the bottle neck if the renderer engine can output more than 500mb/s of data.

I don't think the test is performed in any worthwhile manner, but with laptops, the GPU is usually what bottle necks renders, not SSD.

I didn't bother to check which GPU he has in the 2019, but the 560X isn't all that better than the 370X found in 2015 model.

The 560X in the 2019 model is pretty similar to the 460 found in 2016 model... which wasn't a giant upgrade to begin with.

Vega20 is were is at, and would probably slay the 2015 model.
[doublepost=1563304178][/doublepost]

Can I chime in and talk about things I don't understand? Because this is what this YouTuber does. Its the SSD that throttles the system. you can't export faster than it is able to write the flash storage. He could have used the activity monitor app and would have seen the i9 idling.

Except FCPX does tons of crap on the GPU...

I have been editing on my mini and the i7 didn't even work, while the craptacular 630 HD was so tasked it bogged the whole system down.

 

[profcutter]

Yeah, sounds like a heat throttling issue. There’s only so much performance you can get out of such a small case without radical cooling solutions. I do suspect a better GPU would make a difference.

 

[lJoSquaredl]

1. The test wasn't that stressful on the system, so it doesn't make use of resources available. Lightroom and 1080p ain't nothing, do a 4k 10-20 mins project with tons of effects and layers and transitions then get back to me. Then you see the CPU get a workout and GPU start sweating.

2. FCPX is extremely well optimized. It's not like Adobe Premiere where you have to keep upping the specs to see better performance, it's just always amazingly fast on any Apple product.

3. Throttling to 2.2-2.6ghz, probably less throttling on old models tho still some I'd imagine.

4. He said he specced out the 2015 model which maxed at 2.8ghz if not mistaken. 2.8ghz 4 core vs 2.3ghz 8 core seems unfair, but for such a fast export it probably helped to have the 2.8ghz in that scenario? Not sure but could factor in.

At the end of the day for simple quick stuff like this specs don't matter much. It's when you start doing extremely involved editing, longer edits/exports, or 3D work that things like 8 cores and Vega GPUs come in real handy.

 

[OSX15]

Some people reported thermal throttling in the I9 Macbook pro's from 2019 in this thread. I now have a new Macbook Pro 15" 2.3 GHz / Vega 20 Macbook Pro, and the sustained clock frequency is about 2.9 GHz. However.. It is getting REALLY HOT. I'm talking between 96 and 99 degrees celsius when doing Cinebench R20, Adobe premiere rendering, VMs, etc.

Are others also experiencing this on the 2019 machines? Also can someone with the 2019 check if the area above the touchbar gets really hot? because mine does.

 

[Ploki]

Some people reported thermal throttling in the I9 Macbook pro's from 2019 in this thread. I now have a new Macbook Pro 15" 2.3 GHz / Vega 20 Macbook Pro, and the sustained clock frequency is about 2.9 GHz. However.. It is getting REALLY HOT. I'm talking between 96 and 99 degrees celsius when doing Cinebench R20, Adobe premiere rendering, VMs, etc.

Are others also experiencing this on the 2019 machines? Also can someone with the 2019 check if the area above the touchbar gets really hot? because mine does.

I don't think any mobile mac in the last few years runs cooler than 99 degrees under heavy load.

 

[siroht]

I don't think any mobile mac in the last few years runs cooler than 99 degrees under heavy load.

This is correct.....also my 2.3 ghz I9 8 core when stressed hovers around 2.9ghz which is a great improvement where the 2018 I9 dropped to its based clock speed or even lower when stressed.

I must say that on Windows 10 via bootcamp my sustained speed is around 3.7 ghz when all cores are stressed.

 

[pshufd]

This is correct.....also my 2.3 ghz I9 8 core when stressed hovers around 2.9ghz which is a great improvement where the 2018 I9 dropped to its based clock speed or even lower when stressed.

I must say that on Windows 10 via bootcamp my sustained speed is around 3.7 ghz when all cores are stressed.

I use an external fan or two when I'm expecting high CPU load on my MBP - which isn't very often. It's plugged into a USB port on my monitor but I normally keep the fan speed very low.

 

[siroht]

The 2019 will definitely be faster than the 2015, look at the Geekbench scores. I think the issue with the test in the video is he’s using an external SSD to run the tests, but there would be a bottleneck from the usb 3. If he run the same tests from the internal SSD on each machine I think you would see a bigger difference.

Agreed....I performed the Bruce X video export test on my 2.3 model and it averaged 33 seconds. Now when I connect my eGPU Radeon VII graphic card my export average is 13 seconds.

Do keep in mind that exporting in FCP relies heavily on the GPU, as well as rendering. Now transcoding is pretty much a CPU dependent task so more cores equals better performance.

That’s why I lean towards the exporting to an external drive being the primary reason why he experienced these results.... bottleneck.

 

[Tech_Mac_Man]

Thanks for everyone's input on this...I am yet to see a proper 2015 Macbook Pro (MAXED OUT) vs. 2019 Macbook Pro (MAXED OUT) in some general everyday test. See if it is worth spending all that money on a NEW 2019 Model if the 2015 can handle everything you throw at it...Would love to see some real world test on YouTube....

 

[Ploki]

Thanks for everyone's input on this...I am yet to see a proper 2015 Macbook Pro (MAXED OUT) vs. 2019 Macbook Pro (MAXED OUT) in some general everyday test. See if it is worth spending all that money on a NEW 2019 Model if the 2015 can handle everything you throw at it...Would love to see some real world test on YouTube....

even the base model 13" from 2018 will outperform a maxed out 2015 15", there's really not that much to it.

whether its worth it and you need it, is another question only you can answer

 

[pshufd]

even the base model 13" from 2018 will outperform a maxed out 2015 15", there's really not that much to it.

whether its worth it and you need it, is another question only you can answer

I think that peace of mind is worth a lot. Can you rely 100% on the 2018 13"?

 

[Ploki]

I think that peace of mind is worth a lot. Can you rely 100% on the 2018 13"?

yeah. (actually no)
great computer.
upgraded from 2012 quad. The 2015 is not much different (i used a few)

issue is, Mojave is messy and audio (which is what i do) is messy and can cause issues.

so no, not 100%.

and even audio is not really super problematic, i have one terrible specific issue connected to Mainstage and background services.

 

[ziyuel]

the CPU is super faster!
that is!

 

[BigBoy2018]

the CPU is super faster!
that is!

In geekbench tests, which are bogus.

The 2019 2.3ghz is TURBO BOOSTING to 4.5ghz. The 2015 is turbo boosting to 3.7ghz. The problem with Geekbench is that you'll see all the benefits of that turbo boost in their short tests.

HOWEVER, in real world heavy use (such as exporting videos) that turbo boost will diminish or be completely eliminated, in order to keep the cpu temps down. (Keep in mind, this isn't 'throttling' it's just that under heavy use, the 2019 will operate closer to 2.3ghz and the 2015 will settle on it's base clock of 2.8ghz). Heavy, longer loads will often drop the processor down to it's base clock speed.

In other words, turbo boost is basically TRICKING Geekbench into higher scores that don't often translate into real world use.

Now, I know some will argue "well in short operations turbo boost will be a benefit" ... to which I say, true, but in short terms operations your saving a second, or maybe a fraction of a second - simply because the operation is so short. I'm not sure that's the kind of speed savings you're really gonna brag about: "My computer exports an edited photo in 7 second now and it used to take 8 seconds! Incredible!"

 

[pshufd]

It's nice to have powerful laptops but I wish that Apple had a proper desktop for those things that you need good performance for where you don't need mobility. Basically something in-between the Mac Mini and the Mac Pro in a tower so that cooling isn't an issue and where you could put in a decent GPU without worrying about heat.