Site Tools


Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
rhino:display_performance [2022/10/15]
maryfugier [Render with One Mesh]
rhino:display_performance [2022/10/18] (current)
maryfugier [Task Manager Performance Testing: Not the right tool]
Line 1: Line 1:
-====== Understanding Display Performance ======+====== Understanding Display Performance in Rhino ======
  
 ===== Question: ===== ===== Question: =====
Line 8: Line 8:
 //When I run the 'TurnTable' command and watch the performance tracking in the Task Manager everything is under 50% utilization (with the 2200 card the GPU was maxing out, with the A4000 it hits 44%).  //When I run the 'TurnTable' command and watch the performance tracking in the Task Manager everything is under 50% utilization (with the 2200 card the GPU was maxing out, with the A4000 it hits 44%). 
 However, Raytrace mode and hits 100% performance in Task Manager// However, Raytrace mode and hits 100% performance in Task Manager//
 +{{:rhino:cpu_bottle_neck_01.jpg?500|}}
 ===== Answer: ===== ===== Answer: =====
  
Line 16: Line 16:
  
 Cycles Raytrace mode uses 100% GPU for everything it does, so of course it’s going to max out most of the time. (Turntable is not designed to work with Cycle’s Raytrace mode.) Raytrace is using the GPU 100% even though the display may still seem slow. Cycles Raytrace mode uses 100% GPU for everything it does, so of course it’s going to max out most of the time. (Turntable is not designed to work with Cycle’s Raytrace mode.) Raytrace is using the GPU 100% even though the display may still seem slow.
 +{{:rhino:cpu_bottle_neck_02.jpg?500|}}
 +
 +Size of the file also has nothing to do with display performance, however **all** the geometry must be processed and meshed by the CPU before the GPU can take over. 
  
-Size of the file also has nothing to do with display performance, however all the geometry must be processed and meshed by the CPU before the GPU can take over. This can appear like the display performance is slow, when it is really a bottleneck at the CPU.+**//This can appear like the display performance is slow, when it is really a bottleneck at the CPU.//**
  
 ==== The CPU Bottleneck ==== ==== The CPU Bottleneck ====
Line 35: Line 38:
 This is what is happening when the Render display model reports 27% or similar number for GPU performance. It is not that the GPU is the issue, it is simply waiting for the CPU to catch up. This is what is happening when the Render display model reports 27% or similar number for GPU performance. It is not that the GPU is the issue, it is simply waiting for the CPU to catch up.
  
 +[[https://www.wepc.com/tips/cpu-gpu-bottleneck|{{:rhino:cpu_bottle_neck.jpg?400|}}]]
 ==== Similar Geometry: Instancing in the Display ==== ==== Similar Geometry: Instancing in the Display ====
 If those 100K objects all looked the same, then Rhino could draw all of them in one fell swoop using a single call to the GPU. If those 100K objects all looked the same, then Rhino could draw all of them in one fell swoop using a single call to the GPU.
Line 42: Line 46:
 Obviously, it can be a bit more complicated, since not every object looks the same…but the concept is still the same.  Obviously, it can be a bit more complicated, since not every object looks the same…but the concept is still the same. 
  
-**The GPU will almost never be the bottleneck for Rhino’s display, especially high-end models. Most of the time spent drawing things is done in Rhino’s object and attribute/material management.**+//The GPU will almost never be the bottleneck for Rhino’s display, especially high-end models. Most of the time spent drawing things is done in Rhino’s object and attribute/material management.//
  
 Rhino has tried to improve this over the years by packing as much as possible on the GPU, but there’s still the problem with the CPU  just iterating over large datasets in general.  Rhino has tried to improve this over the years by packing as much as possible on the GPU, but there’s still the problem with the CPU  just iterating over large datasets in general. 
  
-**The typical recommendation for those with these issues is to first improve your CPU (more cores and higher processing speed), add more system memory, then improve your graphics card.** +//**The typical recommendation for those with these issues is to first improve your CPU (more cores and higher processing speed), add more system memory, then improve your graphics card.**//
  
 ==== Mesh First - Then Render ==== ==== Mesh First - Then Render ====
Line 104: Line 108:
  
 === The Race Car Analogy === === The Race Car Analogy ===
-I have a standard Toyota Civic sitting at the starting line.  I can’t “go” until all of my tires are changed, and my tank is filled with gas…+{{:rhino:cpu_bottle_neck_03.jpg?400|}} 
 + 
 +I have a 90's 4-dr sedan sitting at the starting line.  I can’t “go” until all of my tires are changed, and my tank is filled with gas…
  
 It takes my pit crew 8 minutes to get everything ready. And then floor it, and it takes me 15 seconds to reach the finish line… So the overall time it took was 8:15 (8 minutes, 15 seconds). It takes my pit crew 8 minutes to get everything ready. And then floor it, and it takes me 15 seconds to reach the finish line… So the overall time it took was 8:15 (8 minutes, 15 seconds).
  
-I go out and I buy a Ferrari … and get on the starting line…+I go out and I buy an new Italian supercar … and get on the starting line…
 My crew still takes 8 minutes to get everything ready… I then floor it, and it takes, me less than 5 seconds to reach the finish line (300% faster)… The overall time it took was 8:05 (8 minutes, 5 seconds)… My crew still takes 8 minutes to get everything ready… I then floor it, and it takes, me less than 5 seconds to reach the finish line (300% faster)… The overall time it took was 8:05 (8 minutes, 5 seconds)…
  
-The crew is the CPU, and the car is the GPU. The Honda Civic is the Intel and Ferrari is the RXT Quadro+The crew is the CPU, and the car is the GPU. The older sedan is the Intel and supercar is the RXT Quadro
  
 If you really want Rhino to be faster, you will need to first invest in a faster CPU… Then augment that with a better GPU. If you really want Rhino to be faster, you will need to first invest in a faster CPU… Then augment that with a better GPU.
rhino/display_performance.1665810690.txt.gz · Last modified: 2022/10/15 by maryfugier