Gaming on 16 inch MacBook pro

[MacBookGamer]

Thank you. Also going to see what Apple Shows off tomorrow, my luck I purchase the Mac book and they release a new one lol

I purchased the 5600m, but it's not going to arrive until next month. If they release a new one, then I can cancel the order.

 

[Aus-Rotten88]

I have a 2.4/5600M/32GB arriving Wednesday.
Had a 5300 and 5500 before, cant wait to compare.

 

[Chompineer]

I have a 2.4/5600M/32GB arriving Wednesday.
Had a 5300 and 5500 before, cant wait to compare.

Man I’m thinking about it so hard... Have a 5500 right now. How did you end up justifying it?

 

[joseph.s.jones]

Yeah I’m in the same boat debating the 5600M. And I don’t need 32GB RAM now but when paying all that money for the 5600M I start thinking that I can’t take the risk that my needs are going to change in 2-3 years and I’ll be stuck with 16GB RAM. Where do you stop then, do you upgrade the CPU to 2.4? It all gets a bit future proofy when paying this much.

 

[Aus-Rotten88]

Go for the 32, I don't wanna sound like a prick but if you're spending $3500 on a laptop then a extra grand for a better bin processor and extra ram shouldn't matter.

If I can have my light gaming, work, music production, and photo editing done on one Laptop then I'm a happy camper.
The 5500m was almost there with gaming, I expect the 5600 to work perfect for my needs.
The real test is going to be if I can run Tarkov at 60fps with no drops, that would be amazing.
My reaction times are slow now anyways so I could care less about refresh rates. I am not a competitive gamer.
I tried moving to windows with my Aorus 15G, but its not working out.
Airdrop is a must for my work flow, Logic is the best DAW ever IMO.
Audio drivers on windows are trash, blue screen crashes everyday.

 

[joseph.s.jones]

Go for the 32, I don't wanna sound like a prick but if you're spending $3500 on a laptop then a extra grand for a better bin processor and extra ram shouldn't matter.

If I can have my light gaming, work, music production, and photo editing done on one Laptop then I'm a happy camper.
The 5500m was almost there with gaming, I expect the 5600 to work perfect for my needs.
The real test is going to be if I can run Tarkov at 60fps with no drops, that would be amazing.
My reaction times are slow now anyways so I could care less about refresh rates. I am not a competitive gamer.
I tried moving to windows with my Aorus 15G, but its not working out.
Airdrop is a must for my work flow, Logic is the best DAW ever IMO.
Audio drivers on windows are trash, blue screen crashes everyday.

You didn’t sound like a prick at all, it’s a perfectly reasonable perspective and I’ve been back and forth on it a fair bit.

 

[diego9]

The main thing for me is that the 16" fans stayed below 2000RPM during the benchmark.

That's exactly it. Without an eGPU, the MBP's fans get VERY loud if you push it, especially if you push the built-in discrete GPU.

With a 5700 XT eGPU, my 16" acts like nothing is happening while putting out performance to shame any laptop out there. Fans at minimum, feels as cool as ever, no noise. As for the 5700 XT itself, it has 3 fans that are very quiet, and I can't hear them even while running games - in 6K resolution, no less.

With a Pro Display XDR in addition to the eGPU, my MBP turned into a "baby Mac Pro", including the upgradability part.

 

[joseph.s.jones]

That's exactly it. Without an eGPU, the MBP's fans get VERY loud if you push it, especially if you push the built-in discrete GPU.

With a 5700 XT eGPU, my 16" acts like nothing is happening while putting out performance to shame any laptop out there. Fans at minimum, feels as cool as ever, no noise. As for the 5700 XT itself, it has 3 fans that are very quiet, and I can't hear them even while running games - in 6K resolution, no less.

With a Pro Display XDR in addition to the eGPU, my MBP turned into a "baby Mac Pro", including the upgradability part.

Yeah the eGPUs are great for the lack of fan noise. I’m naively hoping that the MBP 16 with turbo boost disabled will actually keep the fans pretty chilled out at 1080p resolution.

 

[diego9]

Whilst the 16" fans stayed below 2000RPM for you, how were the fans on the Razer Core X Chroma? A bit of googling suggests they're ridiculously loud. Did you find them less loud than the MBP 16?

I wonder if the 5600M would actually be able to keep the fans relatively quiet if it was just running a game at 1080p medium graphics. Rather than the YouTube video above where it’s being pushed to its max.

The fans on my 5700 XT in the eGPU are SUPER QUIET. As I said in my previous post, I can't hear them even while playing games. And I'm very particular about noise. I was on the verge of selling my MBP solely because of its fan noise - and then I decided to try an eGPU as a last ditch effort.

My enclosure is a Sonnet Breakaway Box 650. It has a built-in fan that I find useless so I disconnected it from the main board (it's just a standard connector, I just pulled it). I leave the enclosure without its cover so that the graphics card can freely run its fans when it wants to, and push the air far away instead of circulating it inside the enclosure. So when not gaming the entire eGPU (enclosure + card) is silent with no fans running. The card is a Sapphire Nitro+ 5700 XT. The best part: both together cost about the same as what Apple's asking for the 5600M upgrade.

I'm here to tell you that the MBP's discrete GPU is THE source of excess heat and fan noise. There's no question in my mind that the 5600M is the same. With the 5500M I was getting fan noise just from casually browsing the web on the Pro Display XDR. Now with the same display connected to the eGPU, the 5700 XT's triple fans idle-stop until I run a game - they simply won't spin if I'm doing something casual on the computer.
[automerge]1592799139[/automerge]

I’m naively hoping that the MBP 16 with turbo boost disabled will actually keep the fans pretty chilled out at 1080p resolution.

I guarantee you it won't. Meanwhile with an eGPU I'm playing Sonic Racing at full 6K resolution and it's perfectly smooth (the game won't show the frame rate but just one look at the screen and you can tell that it's too smooth to be any less than the display's refresh rate of 60Hz).

 

[joseph.s.jones]

The fans on my 5700 XT in the eGPU are SUPER QUIET. As I said in my previous post, I can't hear them even while playing games. And I'm very particular about noise. I was on the verge of selling my MBP solely because of its fan noise - and then I decided to try an eGPU as a last ditch effort.

I'm here to tell you that the MBP's discrete GPU is THE source of excess heat and fan noise. There's no question in my mind that the 5600M is the same. With the 5500M I was getting fan noise just from casually browsing the web on the Pro Display XDR. Now with the same display connected to the eGPU, the 5700 XT's triple fans idle-stop until I run a game - they simply won't spin if I'm doing something casual on the computer.

This is really encouraging to hear about the eGPU. As for the dGPU casually causing that much heat, it’s a bit surprising. In this benchmark of the 5600M playing the relatively non challenging game CS:GO at max settings on native resolution, the GPU fan is only at 2500 RPM. At 1080P and/or capping the FPS to 60 (rather than the 125-200 FPS it’s getting here) it should be a breeze.

 

[diego9]

As for the dGPU casually causing that much heat, it’s a bit surprising.

Part of it is the Pro Display XDR. It has about 10x the pixels of a 1080p display. So that's a lot of bandwidth to handle for the dGPU. Such high bandwidth translates to heat in the MBP. (If you ever saw a 10gbps network interface up close, you know that those run hot from all the bandwidth they have to handle. A single Pro Display XDR uses 2 to 4 times as much bandwidth just to operate at a standard 60Hz refresh rate.) But if the display is connected to the eGPU, that's a whole different story. The dGPU essentially stays idle while the eGPU is doing all the work.

2500 RPM as seen in the video is just under the threshold where the MBP's fans begin to become an annoyance. My MBP's fans stay at the absolute minimum RPM of around 1820-1850 while playing Sonic Racing. It's the exact same RPM when the computer is idle. With the eGPU, the games are like nothing for the MBP. I have realized now that computer games are 99% GPU and 1% CPU, lol.

Btw, the MBP doesn't have a "GPU fan", it's just one cooling system with 2 fans for all heat generated in the system (CPU, GPU, TB3 controllers, etc.). The RPM shown in the video is for this unified cooling system.

Note: using an eGPU with Windows (Bootcamp) is not supported by Apple and may be tricky to setup. I'm playing macOS games (Apple Arcade).

 

[joseph.s.jones]

Part of it is the Pro Display XDR. It has about 10x the pixels of a 1080p display. So that's a lot of bandwidth to handle for the dGPU. Such high bandwidth translates to heat in the MBP. (If you ever saw a 10gbps network interface up close, you know that those run hot from all the bandwidth they have to handle. A single Pro Display XDR uses 2 to 4 times as much bandwidth just to operate at a standard 60Hz refresh rate.) But if the display is connected to the eGPU, that's a whole different story. The dGPU essentially stays idle while the eGPU is doing all the work.

2500 RPM as seen in the video is just under the threshold where the MBP's fans begin to become an annoyance. My MBP's fans stay at the absolute minimum RPM of around 1820-1850 while playing Sonic Racing. It's the exact same RPM when the computer is idle. With the eGPU, the games are like nothing for the MBP. I have realized now that computer games are 99% GPU and 1% CPU, lol.

Btw, the MBP doesn't have a "GPU fan", it's just one cooling system with 2 fans for all heat generated in the system (CPU, GPU, TB3 controllers, etc.). The RPM shown in the video is for this unified cooling system.

Note: using an eGPU with Windows (Bootcamp) is not supported by Apple and may be tricky to setup. I'm playing macOS games (Apple Arcade).

Yeah you’re right, the eGPU is always going to win from a noise perspective. I just found it kind of impressive that the 5600M could run it at max settings, native res and pull those FPS numbers without the fans really spinning up. Admittedly CS:GO could run on a potato, so it’s probably not that impressive all things considered. But it makes me wonder how something like CIV 6 or Warhammer Total War 2 would run at 1080P mid-high 🤷🏻‍♂️

Thanks I didn’t know that about the cooling system. Yeah I had heard about the eGPUs being a bit of a pain to get working sometimes. And that an NVIDIA card was recommended for bootcamp but an AMD card for MacOS.

 

[JPizzzle]

Specing my “gaming” setup now and Wondering thoughts on 2.3 vs 2.4 cpu and ssd size. I’d love to be able to get 1tb, but some of these games are huge. Warzone is 175gb alone.

 

[Aus-Rotten88]

Went with 2TB, a quality tb3 1tb ssd would cost the same as the upgrade.
COD always crashed off my samsung t5 for whatever reason.
Rumors are silicon lottery wise the 2.4 is a better bin processor.
I just hope the drivers are updated somewhat on the 5600m, 5500m was a total mess at launch in bootcamp.

 

[imrazor]

I've got a 4GB 5500m and noticed little difference in benchmarks between it and the 5300M. I attributed this to the fact that both GPUs have a 50w power limit, even though AMD specs show that the 5500M is capable of using much more wattage. No doubt this is due to thermal performance of the GPU cooler. But it makes me a little dubious that the 5600M is going to be substantially better than the 5300M/5500M.

 

[Chompineer]

I've got a 4GB 5500m and noticed little difference in benchmarks between it and the 5300M. I attributed this to the fact that both GPUs have a 50w power limit, even though AMD specs show that the 5500M is capable of using much more wattage. No doubt this is due to thermal performance of the GPU cooler. But it makes me a little dubious that the 5600M is going to be substantially better than the 5300M/5500M.

The 5600m uses far less power as it relies on a large amount of cores (40CU vs 22CU on the 5500M) at lower clock speeds and high efficiency HBM2 memory to improve performance.

It takes a lot of wattage to gain a little performance when it comes to clock speeds. Having a ton of cores at low clock speeds is a great way to improve performance and keep power draw down.

The giant 40CU die (same as the full size 5700XT) and HBM2 memory are very expensive ways to increase performance, but it does allow you to stay in a 50w envelope.

 

[CibbySF]

I have a 2.4/5600M/32GB arriving Wednesday.
Had a 5300 and 5500 before, cant wait to compare.

You might be one of the first w a 5600 arriving. I'm so curious to see if it draws less Watts powering an external display w lid open. Will you please report what you experience?

on your 5300/5500 builds, from a fan noise POV did you find the base model better? From what I can tell online fans ramp up less quickly on the base. I have an i9 5500 but in my return window so trying to figure out the best next config to try

 

[joseph.s.jones]

You might be one of the first w a 5600 arriving. I'm so curious to see if it draws less Watts powering an external display w lid open. Will you please report what you experience?

on your 5300/5500 builds, from a fan noise POV did you find the base model better? From what I can tell online fans ramp up less quickly on the base. I have an i9 5500 but in my return window so trying to figure out the best next config to try

I’m also interested in this.

 

[CibbySF]

Got this from Max Tech. They can't yet measure the wattage of the Radeon High Side, iStats isn't reporting it. But he sent me his temps while browsing for an hour and plugged into an Apple 6K XDR

 

[Aus-Rotten88]

Just ran test FireStrike test, The only thing I did was disable Turbo. 5500M was running BootCamp Red Drivers.

9 094 with AMD Radeon Pro 5500M(1x) and Intel Core i9-9880H Processor
Graphics Score 10639
Physics Score 11355
Combined Score 3810


12 840 with 5600M and Intel Core i9-9980HK Processor
Graphics Score 15881
Physics Score 12911
Combined Score 5253

My friends gx501 with a 1080 max Q scores around 12500.

Here is my Aorus 15G with a 2070 Super


17 492 with NVIDIA GeForce RTX 2070 SUPER (Notebook)(1x) and Intel Core i7-10875H
Graphics Score 19670
Physics Score 19377
Combined Score 8853

Here is the 5600m with Turbo on, fans at 100%

13 035 with 5600M and Intel Core i9-9980HK Processor
Graphics Score 15503
Physics Score 15245
Combined Score 5407

Battery drain was 3% during the test so you'll have to disable Turbo if you plan on gaming with this thing. I'm starting to download a couple games right now but overall I am very impressed with the results. I think this is the gaming laptop that was rumored back in the beginning of the year.

The only external device I have is my Vizio TV that is 4k 60hz or 1080 120hz.

 

[Aus-Rotten88]

Just played a couple rounds of Valorant,
Aorus 15G at 1080 gets around 220FPS vs 160FPS on the 5600M @ 3072x1920
Response time is slow on the Mac screen and being 60hz it is what it is but for whatever reason performance was the same (mid tier)

 

[Hjupter Cerrud]

Just played a couple rounds of Valorant,
Aorus 15G at 1080 gets around 220FPS vs 160FPS on the 5600M @ 3072x1920
Response time is slow on the Mac screen and being 60hz it is what it is but for whatever reason performance was the same (mid tier)

That's a pretty good results 160FPS on 3k vs 220FPS on 1080p, you should try 1080 on the Mac just to see how much does improve the FPS. Damn it this makes me want to sell my 1 month old MBP 16 XD.

 

[Aus-Rotten88]

1200 is the sweet spot with 200 frames, 1080 for whatever reason isn't stable during a match. Bounces from 130-240.
Just for laughs I tested 1080 during a spike rush defuse practice round. I was getting 400 frames - I'm not sure how that's possible.
Testing Tarkov now.

 

[Aus-Rotten88]

60FPS on Tarkov Medium setting 1920x1200, very respectable considering its not well optimized on any system.
5500M would get 50FPS Low settings on 1050.
So far I'm loving this card. Can't wait for updated drivers from BootCampDrivers.

 

[alsid02]

60FPS on Tarkov Medium setting 1920x1200, very respectable considering its not well optimized on any system.
5500M would get 50FPS Low settings on 1050.
So far I'm loving this card. Can't wait for updated drivers from BootCampDrivers.

That's great, also looking to pull the trigger on this (2.3/5600M/32GB). Can you please confirm if it still has the issue of overheating/loud fan noise when connected to external display?