Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Sunrunner

macrumors 6502a
Original poster
Nov 27, 2003
600
2
Now that we have these shiny new Mac Pros, looking for a game to use as a stress test. The goal is to run the most visually stunning environment possible in highest settings; anybody have a good recommendation? I have tried a few on-hand test articles (Civ VI and others) without the MP getting above ~10%...
 
Now that we have these shiny new Mac Pros, looking for a game to use as a stress test. The goal is to run the most visually stunning environment possible in highest settings; anybody have a good recommendation? I have tried a few on-hand test articles (Civ VI and others) without the MP getting above ~10%...
Using Terminal test memory pressure - 'memory_pressure -l critical'. It used to crash the system a few moons ago, but Apple fixed it. 👍

Running this on my late 2016 15" rMBP13,3 waiting for my MP7,1.... :) Use Ctl C to terminate it.
 

Attachments

  • Screen Shot 2019-12-24 at 7.56.28 AM.png
    Screen Shot 2019-12-24 at 7.56.28 AM.png
    935.4 KB · Views: 292
  • Screen Shot 2019-12-24 at 8.05.55 AM.png
    Screen Shot 2019-12-24 at 8.05.55 AM.png
    228.8 KB · Views: 263
Last edited:
Now that we have these shiny new Mac Pros, looking for a game to use as a stress test. The goal is to run the most visually stunning environment possible in highest settings; anybody have a good recommendation? I have tried a few on-hand test articles (Civ VI and others) without the MP getting above ~10%...
Saw a Youtuber running Forza 4 and Gears 5 at 6K Ultra setting on a XDR Display it’s was making out the CPU and GPU but system remained silent and at 60 FPS
 
In MacOS, I'd recommend Shadow of the Tomb Raider. Looks beautiful and will definitely push your system. I ran the built-in benchmark at 4K on my Radeon VII on my new 7,1 and got 49 FPS average.
 
  • Like
Reactions: h9826790
Now that we have these shiny new Mac Pros, looking for a game to use as a stress test. The goal is to run the most visually stunning environment possible in highest settings; anybody have a good recommendation? I have tried a few on-hand test articles (Civ VI and others) without the MP getting above ~10%...
Run a few instances of the Stockfish.app (https://stockfishchess.org). This will ramp up a whole lot of things.... it will also show how well the worker-bees/cores can burn the resources up.
 
Can you test GPU temps under heavy load and compare to any PC review of the same GPU? The back plate of the GPU in the Mac Pro has only 1mm space between it and the PSU. GPUs need airflow all around them on top and bottom.

Apple did the same mistake as the 5,1 with the motherboard facing the wrong direction so that the back plate is on the bottom and is almost in contact with a nearby metal surface.
 
Apple did the same mistake as the 5,1 with the motherboard facing the wrong direction so that the back plate is on the bottom and is almost in contact with a nearby metal surface.

You mean the graphic card?

On 5,1, there is a plastic plate on the CPU cage. The graphic card can lie on it, won't touch any metal.

And I tried to install graphic cards (HD7950, R9 380, 1080Ti, RX580, Radeon VII) in different slots (on 5,1). All run coolest in slot 1, the bottom slot that you said no air to flow on the back side. So, allow airflow on both upper and lower side of the graphic card won't provide better cooling.

So, there is no mistake from Apple on this matter. Do you have any prove to show us that's a mistake? Your theory doesn't fit my observations.
 
  • Like
Reactions: thisisnotmyname
You mean the graphic card?

On 5,1, there is a plastic plate on the CPU cage. The graphic card can lie on it, won't touch any metal.

And I tried to install graphic cards (HD7950, R9 380, 1080Ti, RX580, Radeon VII) in different slots (on 5,1). All run coolest in slot 1, the bottom slot that you said no air to flow on the back side. So, allow airflow on both upper and lower side of the graphic card won't provide better cooling.

So, there is no mistake from Apple on this matter. Do you have any prove to show us that's a mistake? Your theory doesn't fit my observations.

That's why I'm asking for temps. On some GPUs the backplate gets hot under prolonged load and the heat has to flow upwards. You have to compare to a PC. I saw the 5,1 with GTX 980 and 1080 was about 10 degrees higher than a PC just because the backplate was facing the wrong direction, and also because the 5,1 has a very small central compartment.

85 degrees in the 5,1. 75 degrees in the PC. Same load test.

Heat rises obviously, that's why graphics cards are designed with the plate on the upper side.
 
Heat rises obviously, that's why graphics cards are designed with the plate on the upper side.

I don't think so. AFAIK, that "reverted" thing is just because when introduce PCI standard, the engineers decided to revert the slot direction, which allow the PCI slots share the same space with the ISA slots. It's nothing about "heat rises".
main-qimg-44d7c4b21c3db1c2444b1b935b1bab2e.jpg



You have to compare to a PC.

So, now is not about card's direction, or may touch the metal, etc. But compare the cases' cooling efficiency?

Then may be we should compare the cMP to a 10 years old PC case which can accomodate two CPUs and 4 internal HDD at that size.

Compare a 10 years old Mac Pro with a modern gaming computer's case, then conclude that "Apple make a mistake" doesn't make any sense to me.

I saw the 5,1 with GTX 980 and 1080 was about 10 degrees higher than a PC just because the backplate was facing the wrong direction

This is easy to test, just turn the cMP upside down, then the graphic card should run 10C cooler. And we can install the card in slot 2 to make sure air can flow through both side. AFAIK, it won't help that much.

All we know is just cMP's PCIe compartment cooling efficiency isn't as good as modern gaming computer (mid to full size tower, not those mini gaming PC). We should not conclude that as "the card facing wrong direction", or "Apple made a mistake", etc. There is simply not enough data to give out these conclusions.
 
  • Like
Reactions: thisisnotmyname
I believe the ultimate test is to Bootcamp into Windows 10 and run GTA V on the Mac Pro. GTA V bakes even the most heavily cooled PCs on Max settings, especially in the areas of the game where there is grass and foliage. It would be impressive if the Mac Pro can survive GTA V on max settings.

Who's willing to take this challenge?
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.