How fast is your Inventor PC really?

[Raider_71]

Hi guys,

We have had to do some testing on a bunch of Inventor PC's recently to determine which of the PC's needs to be replaced. Obviously we needed to find out which of the PC's are the worst eprformers as there was only budget to replace 50% of the design PC's. So we thought the Darwin theory will come in handy right... 🙂

Anyways I started searching on the net for toppics on how to benchmark an Inventor PC. Then I thought whats the point of using gaming benchmark tools because Inventor is not a game and there are more aspects than just graphics performance when it comes to percieved performance on an Inventor PC right.

So we decided to create our own Inventor benchmark tool which tests various aspects of an Inventor PC to give us an overview of our PC's performance. This then helped us make a decision as to whcih pc's to replace.

We have made the tool available free of charge to anyone interested in checking how their PC stacks up to their peers or friends. 🙂

Please download it here and post your results here as well if you want. Would be interesting to see what beast workstations are out there.

I would like to say thanks to Kirk #karthur1, for helping in testing the app.

Please feel free to send any suggestions our way. There is an email link in the app.

Download and Install

The application will work with Inventor 2014 to 2016 only.

IMPORTANT: After installation there will be an Inventor Bench icon on your desktop that looks like this: 

My resluts:

HP Elitebook 8560w with an SSD upgrade.

[Neil_Cross]

@Raider_71 Apparently the API call for gathering the polygon/face count is SurfaceBody.GetExistingFacets.  One of the new features of 2018 causes an effect on this but obvs I can't talk about that in public in any specific detail, but you'll probably know what I mean.

All this info should be shown in the debug stats option but as you've probably seen it doesn't work, it's all just zero's for polygon counts so they're stepping up an effort to get that fixed, I doubt there'll be any urgency on that though as it's a hidden registry feature which not many people care about.

[SBix26]

---

@brotherkennyh wrote:

I just tried this.

I created a model. A Cube with a bunch of polygons, fillets, shells etc.

I checked my CPU/GPU load while 1 part was in an assembly while panning around.

I patterned the part 100 times, then 3375, then 8000 times. In each case there was no increase in memory footprint or CPU/GPU load etc.

I did create some screenshots showing the difference, but you cant even tell the difference in the performance results so there is little point showing them.

---

Would patterning at different angles (curved paths) give the needed variety?  Then the instances would not be visually identical and would need to be rendered individually.

Sam B

Inventor Professional 2017.3.1
Vault Workgroup 2017.0.1
Windows 7 Enterprise 64-bit, SP1
Inventor Certified Professional

[Neil_Cross]

Not a bad idea to test that.  I suppose it depends if the back end is intelligent enough to re-use the asset as a detected duplicate object rather than how it visually appears on screen 

[Raider_71]

Hi guys,

Here is something to play with. It's a simple script which will generate a random assy with specified number of rows and levels of random and unique parts. Each part generated is 100% unique.

Each part also contains some visible work planes and 2D sketches which further complicates things for the graphics card. This can be switched off obviously if you want to check what the effect would be without it.

Give it a go and let me know what you guys find.

https://dl.dropboxusercontent.com/u/30964827/Temp/RandomAssy.zip

Just keep the two files inside the zip file together in a folder then it should work fine.

Cheers

[mmaes]

@Raider_71 IS there a way to do this with earlier versions?  I am only on 2016

[Raider_71]

Hi @mmaes

Oh darn I was hoping that that would not happen...

I built the seed part in 2017 so the only way is to reconstruct it in 2016. I will see if I get time over the weekend. I need to do it anyways as I would like to include something like this in the next InventorBench tool but also with assembly constraints.

Sorry for that...

[mmaes]

out of curiosity I decided to run the bench test at 4k resolution next to 1080

I am actually surprised it did such a number on the score

[w_meerkerk]

My new laptop:

Any way to run the test on the dedicated Graphics card? This is on my IGP 😞

[Neil_Cross]

It probably is running Inventor on the dedicated GPU, the test just reports the iGPU as being the primary adapter.  You can go into Intels management software or it might be the Nvidia control panel and force it to make sure the proper GPU is the primary first choice.  The test might still report on the iGPU though.

[w_meerkerk]

Ah ok, that's clear. Indeed the nVidia control panel says Inventor is running on the Quadro.

[Neil_Cross]

Which mobile Quadro is it? 

[w_meerkerk]

Quadro M2000M 4gb

[Neil_Cross]

That's absolutely fine, it's been conclusively proven already that the power of the GPU doesn't make any difference in Inventor, it's mostly all about the VRAM and you've got plenty with that.

If you're not convinced on that, check this out:

[mmaes]

@Raider_71 the assembly tool you sent me over the weekend did work.  Below is a screen shot of the time.  I don't know if the time even means much at this point though.

[Anonymous]

---

@Anonymous wrote:

---

@brotherkennyh wrote:

 I would be interested to hear about how SW performs with large assemblies like the ones mentioned here. Nothing like a bit of competition to motivate Autodesk.

This i can help with! give me a couple of days though.

---

i took a run at this yesterday after i had finished all my other work and it took nearly 2hrs just to open half the model. (i remember why i hated solidworks so much now.) i will have another crack on one of the sub assemblies when i get a chance in the next few days. i will also try getting some proper system monitoring tools before i try that to see where and why solidworks is taking so long to open things.

[Raider_71]

Hi @mmaes

For now, the tool is just for your own use and to be able to generate large assemblies with loads of unique parts if you would like to check the effect of this on your system's graphics card.

I will be building in an assembly test sequence which will operate similar to this but will also include constraints and constraint editing.

Cheers

[Anonymous]

Hi,

tested with CPU i7-7700K @ 4,8 GHz

1. regular test

2. a trick with the window

3. slow it down with the user panel

2. The second test isn't a fake but it was done with a little trick:

- start Inventor

- cover the Inventor window with an other window. But not as a full window.

- start Inventor bench

The Inventor window has to be hidden during the test . Sometimes it takes several test to get minimum graphic times.

3 . The third test:

- start inventor

- start new part

- pull a panel from the ribbon menu in the graphic window (see the picture below)

- restart inventor

- start the Inventor bench

 

A comment to the test 1 and 2:

If you compare the modeling time in the first (8,98 s) an second (6,49 s)  picture there is a approx. 40% difference.

Is this the impact of the graphic card on the modeling time?

I'm going to do some tests with importing large Step files in a hidden window.

[mmaes]

I'm not sure the sense in that sort of testing but I do believe the GPU "may" have some impact on modeling time. Yesterday I overclocked my GPU and forced it to stay in boost mode then ran the bench test. Results were 13.5x IPI and faster modeling time.

[Neil_Cross]

The GPU doesn't impact on modelling compute times, I had it confirmed by the lead principal guy on the graphics team that there's no GPU acceleration, CUDA or OpenCL, nothing in Inventor is accelerated on the GPU not even the output to the monitor.

I would argue the variation in compute times will be down to CPU load at the time of the test.  I've been doing manual tests none stop for the past 3 days (video uploading to youtube now) and even in manual tests, there's massive variations in how long it takes Inventor to perform a task purely due to the majority of Inventor being CPU dependent, and the CPU being pulled off to do other things during testing, even insignificant background processes can have a knock on effect.  For example I did a 1000 part shrinkwrap that took 80 seconds, I did the exact same thing again immediately afterwards on the same assembly and the exact same task took 100+ seconds, then again it took 85, it's just PC load at testing time, sporadic bursts of CPU resource going to system services can easily knock a second onto a bench test.

[Anonymous]

Hi,

@Neil_Cross

What I meant by this comment "Is this the impact of the graphic card on the modeling time?" is the question:

Is the graphic window slowing the modeling down?

@mmaes

I agree that there is no practical use for a test in a hidden window.

But if there is a "hidden power" in Inventor, that can speed up for example the importing of data or the deriving of a large idw, than I would like to use that "power".