Poor CPU performance with eGPU connected?
 
Notifications
Clear all

Poor CPU performance with eGPU connected?  

  RSS

guatemalajoe
(@guatemalajoe)
New Member
Joined: 2 years ago
 

hi all,

Has anyone experienced poor CPU performance when an eGPU is connected?

I've just picked up a TB3 enclosure and in the main everything on the GPU side looks all good (compute performance, TB speed etc.).  Where things get strange is that with the enclosure connected CPU performance seems to drop, or stay pegged in a low performance/thermal mode. This is most visible in games - there is a clear CPU bottleneck and sub-standard performance.

DOOM 2016 with its performance metrics provides a good way to illustrate this; with the eGPU disconnected I see CPU performance in the realm of 8-16ms (with variation but enough to deliver 60FPS) but with the enclosure connected this value is pegged above 30ms, effectively limiting my frame rate to <30FPS.

My setup:

  • 2017 Macbook Pro 13 (non touchbar), i7-7660U, 16GB RAM
  • HP Omen Enclosure connected to the lower left TB3 Port
  • AMD R9 290x
  • Windows 10 Pro 1803 running under bootcamp
  • Latest AMD Drivers
  • Latest Intel iGPU drivers
  • Booting via rEFInd with flag in config to keep iGPU alive
  • Disabled iSight PCIE to free up bandwidth

What I've attempted:

  • Remove both AMD and Intel drivers and re-install using DDU
  • Use alternate Thunderbolt port
  • Use throttlestop to force Turbo mode on CPU
  • Ensure power management is in high performance mode

Short of reinstalling windows 10 (a PITA on a Macbook) I'm stumped. On paper my old desktop machine (i5-750) should have similar CPU performance to the newer i7-7660 but in gaming (my use case for the dock) its miles off...

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
zmyslony
(@zmyslony)
Eminent Member
Joined: 2 years ago
 

I guess the problem is not with CPU, it's with GPU. Modern games mainly use graphics. On my setup MacBook doesn't even turn on the fans when I play games (since using eGPU).

Try turning off Intel Graphics in Device Manager and using external display.

13" MacBook Pro + MSI GeForce RTX 2080 Duke (Razer Core X)


ReplyQuote
(@alaxend0l_games)
Active Member
Joined: 2 years ago
 

I know this is a very late bump, but I'm having the same issues and I can't figure out why. I'm using a MacBook Pro 16" with the i9 9880h and the 5500m with 4gb of vram, along with a Razer Core V2 and a Radeon VII. I get way, way better performance in DOOM 2016 without the eGPU, and terrible performance with it. In fact, even if I limit the game to using fewer cores, the performance is exactly the same. Is there any way to improve CPU performance? I've tried disabling the intel graphics in Device Manager, but that didn't help at all. Any help would be appreciated!

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Mini i5
(@mini-i5)
Reputable Member
Joined: 7 months ago
 

@alaxend0l_games

Running at what quality and resolution? On what display? Using Vulkan or Open GL? With asynchronous compute on or off? 

2018 Mac Mini i5 / 6-core 3.0-4.1 Ghz / 32GB / 256GB / Intel UHD 630 / PowerColor Gaming Station / PowerColor dual-fan RX5700 XT / Mac OS 10.15.4 Catalina / Windows 1809 / LG 27" 4K display


ReplyQuote
(@alaxend0l_games)
Active Member
Joined: 2 years ago
 

@mini-i5

Basically any resolution is CPU limited, at least within reason. Quality settings don't matter either, low and ultra are the same. Both internal and external monitors are affected. Both OpenGL and Vulkan are affected too, though OpenGL is worse. And what is asynchronous compute?

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Mini i5
(@mini-i5)
Reputable Member
Joined: 7 months ago
 

@alaxend0l_games

Doom 2016 runs at a comparatively high FPS rate - even at 4K ultra settings. A high FPS rate bottlenecks TB3.

Try testing the VII by running the superposition 4K benchmark with the external display connected to the razor core v2.

https://benchmark.unigine.com/superposition

 

2018 Mac Mini i5 / 6-core 3.0-4.1 Ghz / 32GB / 256GB / Intel UHD 630 / PowerColor Gaming Station / PowerColor dual-fan RX5700 XT / Mac OS 10.15.4 Catalina / Windows 1809 / LG 27" 4K display


ReplyQuote
(@alaxend0l_games)
Active Member
Joined: 2 years ago
 

@mini-i5

I'll test Superposition in a minute, but using Doom 2016's built-in performance statistics with the eGPU plugged in, the GPU is never the issue, it's always the CPU unless I'm looking at the corner of a wall or something. Without the eGPU, I'm usually able to get about 120FPS, but with the eGPU, I can't get above 60 unless I'm looking in a corner.

I also gave Titanfall 2 a shot, and I had some of the same issues. In the opening area, looking away from a hallway, I can get 300FPS, but looking at the hallway, I get 30FPS or worse. I don't think this is normal, is it?

More info if it helps: I'm also using apple_set_os v0.5 in order to make my eGPU work at all in Bootcamp. Could that be causing some bottlenecking issues? https://egpu.io/forums/mac-setup/macbook-pro-16-egpu-hardlock-and-bootloop-issues/#post-72736

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Mini i5
(@mini-i5)
Reputable Member
Joined: 7 months ago
 

@alaxend0l_games

The superposition compute benchmark will get away from all the potential variables introduced by graphic workloads in specific scenes in your gaming collection.

Basically, the superposition score will tell you if your overall setup is running properly. So it’s a good step back and look at the big-picture kind of thing.

This post was modified 2 months ago

2018 Mac Mini i5 / 6-core 3.0-4.1 Ghz / 32GB / 256GB / Intel UHD 630 / PowerColor Gaming Station / PowerColor dual-fan RX5700 XT / Mac OS 10.15.4 Catalina / Windows 1809 / LG 27" 4K display


ReplyQuote
(@alaxend0l_games)
Active Member
Joined: 2 years ago
 

@mini-i5

I've attached a screenshot of the superposition benchmark. This seems to be a fairly reasonable score for the Radeon VII.

 Annotation 2020 02 06 100801

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Mini i5
(@mini-i5)
Reputable Member
Joined: 7 months ago
 

Great, so no problems with your overall setup. 

Gaming can require a higher level of real-time data feeding than benchmarking does. As you’ve found, specific scenes’ real-time data needs saturate the available bandwidth of TB.

Which is likely why you’re seeing better performance with the internal 5500m.

There’s some detailed explanation of this by @p-mac in another thread:

https://egpu.io/forums/mac-setup/pcie-slot-dgpu-vs-thunderbolt-3-egpu-internal-display-test/paged/3/

 

This post was modified 2 months ago

2018 Mac Mini i5 / 6-core 3.0-4.1 Ghz / 32GB / 256GB / Intel UHD 630 / PowerColor Gaming Station / PowerColor dual-fan RX5700 XT / Mac OS 10.15.4 Catalina / Windows 1809 / LG 27" 4K display


ReplyQuote
Squuiid
(@squuiid)
Eminent Member
Joined: 9 months ago
 
Posted by: @guatemalajoe

hi all,

Has anyone experienced poor CPU performance when an eGPU is connected?

I've just picked up a TB3 enclosure and in the main everything on the GPU side looks all good (compute performance, TB speed etc.).  Where things get strange is that with the enclosure connected CPU performance seems to drop, or stay pegged in a low performance/thermal mode. This is most visible in games - there is a clear CPU bottleneck and sub-standard performance.

DOOM 2016 with its performance metrics provides a good way to illustrate this; with the eGPU disconnected I see CPU performance in the realm of 8-16ms (with variation but enough to deliver 60FPS) but with the enclosure connected this value is pegged above 30ms, effectively limiting my frame rate to <30FPS.

My setup:

  • 2017 Macbook Pro 13 (non touchbar), i7-7660U, 16GB RAM
  • HP Omen Enclosure connected to the lower left TB3 Port
  • AMD R9 290x
  • Windows 10 Pro 1803 running under bootcamp
  • Latest AMD Drivers
  • Latest Intel iGPU drivers
  • Booting via rEFInd with flag in config to keep iGPU alive
  • Disabled iSight PCIE to free up bandwidth

 

What I've attempted:

  • Remove both AMD and Intel drivers and re-install using DDU
  • Use alternate Thunderbolt port
  • Use throttlestop to force Turbo mode on CPU
  • Ensure power management is in high performance mode

 

Short of reinstalling windows 10 (a PITA on a Macbook) I'm stumped. On paper my old desktop machine (i5-750) should have similar CPU performance to the newer i7-7660 but in gaming (my use case for the dock) its miles off...

This is a bug in Doom 2016 and AMD unfortunately. I’ve spent hours and hours on it and could not find a fix. On NVIDIA it ran perfectly.

I switched from a GTX 1080 to a RX5700XT, using the same CPU/machine and Doom 2016 CPU performance went to hell (no pun intended 😬). It’s almost unplayable no matter what resolution is used. CPU usage just maxes out and impacts frame rate.

 

If anyone finds a fix that would be awesome as I have given up. I suspect it is garbage AMD drivers though.

 

Mac mini 2018, i7 3.2GHz 6c, 32GB RAM, 1TB SSD, 10GbE, Sonnet eGFX 350W (Noctua NF-A12x25 ULN) + PowerColor RX 5700 XT 8GB, TEKQ Rapide TB3 1TB SSD, macOS Catalina, Win10 Pro
Mac Pro 2010, X5690 3.46GHz 6c, 64GB RAM, Samsung SM951 512GB, Sapphire Pulse RX 580 8GB, Intel X540-T1 10GbE, Allegro Pro USB 3.0, AC/BT 4.2, macOS Mojave


ReplyQuote
guatemalajoe
(@guatemalajoe)
New Member
Joined: 2 years ago
 
Posted by: @alaxend0l_games

I know this is a very late bump, but I'm having the same issues and I can't figure out why. I'm using a MacBook Pro 16" with the i9 9880h and the 5500m with 4gb of vram, along with a Razer Core V2 and a Radeon VII. I get way, way better performance in DOOM 2016 without the eGPU, and terrible performance with it. In fact, even if I limit the game to using fewer cores, the performance is exactly the same. Is there any way to improve CPU performance? I've tried disabling the intel graphics in Device Manager, but that didn't help at all. Any help would be appreciated!

Damn, thats a shame that you're seeing the same issue on a more potent machine - I was considering the 6C/12T Macbook. In my use case I saw severe performance degradation in almost any game that required anything beyond a very light CPU.

 

Left 4 Dead ran well but on seeing the Doom frame times I assumed a hardware bug and ditched the setup. Another game you could try to validate the CPU issue would be Battlefield 3/4/1 - these have some good performance metrics that show CPU and GPU frame times as a graph (be sure to turn vsync off). Games that leverage Vulkan may be worth a look too - Wolfenstein SOTC/BF have a Vulkan renderers that reduce CPU overhead.

 

With your more recent example I think its indicative that more cores/threads wont solve this and as it's probably driven by some form of software issue. This could be any host of things - graphics API (d3d, vulkan, oGL) graphics drivers, TB driver, chipset driver, even the audio driver causing this additional cpu latency/poor performance.

 

With this being an edge case on top of an edge case I don't think it will be solved without some serious effort. It's a shame that Nvidia hardware is unsupported in Mac OS as this issue kills the viability of an eGPU setup IMO.

 

If you are willing to go nVidia there are examples online of much lesser hardware working as expected. The example below shows fairly consistent sub 10ms CPU frame times on a MacBook Air with Thunderbolt 1, effectively eliminating the idea that this is a TB bandwidth issue.

 

 

 

 

 

 

This post was modified 2 months ago

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
(@alaxend0l_games)
Active Member
Joined: 2 years ago
 

I've done some more testing, and I think it's an issue with Vega and Navi cards. Doom runs just fine on my RX 580, but using an RX 5700 XT, the game just chokes, just like with the Radeon VII. It's not a bandwidth issue, the game runs better on the RX 580 than the RX 5700 XT. It seems like an issue with AMD's drivers for Vega and Navi.

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Squuiid
(@squuiid)
Eminent Member
Joined: 9 months ago
 
Posted by: @alaxend0l_games

I've done some more testing, and I think it's an issue with Vega and Navi cards. Doom runs just fine on my RX 580, but using an RX 5700 XT, the game just chokes, just like with the Radeon VII. It's not a bandwidth issue, the game runs better on the RX 580 than the RX 5700 XT. It seems like an issue with AMD's drivers for Vega and Navi.

It’s 100% an AMD driver issue. As previously stated, running a GTX 1080 in the exact same eGPU, Mac mini, OS version, everything the same and Doom 2016 runs beautifully.

Throw in the 5700XT and CPU maxes and frame rates tank.

Garbage AMD drivers, yet again.

Mac mini 2018, i7 3.2GHz 6c, 32GB RAM, 1TB SSD, 10GbE, Sonnet eGFX 350W (Noctua NF-A12x25 ULN) + PowerColor RX 5700 XT 8GB, TEKQ Rapide TB3 1TB SSD, macOS Catalina, Win10 Pro
Mac Pro 2010, X5690 3.46GHz 6c, 64GB RAM, Samsung SM951 512GB, Sapphire Pulse RX 580 8GB, Intel X540-T1 10GbE, Allegro Pro USB 3.0, AC/BT 4.2, macOS Mojave


ReplyQuote