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]

Dunno, something must have been going on though as your save time is off the charts for a SSD.  So far the typical save time for a SSD has been around 5-6 seconds, yours was 18+, so there must be some background stuff going on with your PC at the time the test was done

[Raider_71]

Hi @Neil_Cross

InventorBench uses a very simple method to calculate the FPS. Its actually using exactly the same method as the old VB 6 app from a few years ago. But basically a camera gets driven through lets say 500 points. The total time it takes for the model to be taken through those points is recorded and then Frames per second is calculated from that. So the values InventorBench shows will probably not be the same as an external program but its calculated and indicative of your Inventor graphics performance.

Hope this helps.

Please let me know if you have any further questions.

Cheers

[salariua]

---

@Anonymous wrote:

I would REALLY like to know how @Anonymous got his K4200 in the range I'm now showing for the R9 380.  

---

I was wondering the same thing. I know the K4200 are slightly better than the K5000 but by a small margin and the 3x better results don't make any sense. To my relief I am having same results as Curtis on his k5000 units otherwise I wold have gone mad searching for the driver/hardware glitch.

I have compared switching themes from Aero to Classic and switching windows visual settings to performance and it didn't change a bit. No difference whatsoever.

[Anonymous]

@niel_cross I didn't have anything running in back ground. The first modeling part didn't have any images show up just what it was doing in the browser. Also noticed some issues while trying to constrain parts together now.

[Ray_Feiler]

Updated for IPI

[Anonymous]

New update.

[AlexFielder]

This is a good score, correct?

[Neil_Cross]

Very good, that Skylake is prooving to be the doggies danglers for Inventor on this benchmark test.

Curious to know how the overall score is worked out, you're at 11.9 and mine was 11.  My scores where slightly higher in some areas but your's better by fractions of a second in most other tests, curious that made it nearly 1 point higher.  Really doesn't matter though your overall scores are better, I think that puts you in No.1 position at this point! Game on!

[JohnHolder]

Here is mine with the extensive test

[Neil_Cross]

Good scores there John but I think the hard disk is slowing you down across the board, you've got a mechanical Seagate Barracuda 7200rpm and it's adding a few seconds onto most of the benchmarks.  If you swapped that out for even just a cheap entry level SSD you'd have a noticeably faster and more productive PC for doing almost anything, and that would filter through into Inventor use

[JohnHolder]

Yeah my old computer had duel SSD in raid 0 and it was blazing fast. When that crashed it hurt. I would love to pop in another, just need a few projects to finish, then I can talk to the boss again. We just setup an office - buying a few PC's and all the bells and whistles we decided to forgo the SSD's for now. 8(

[Raider_71]

Hi @Neil_Cross

Not sure which of your scores were higher compared to @AlexFielder scores?? According to what I can see his is better in every category unless I am missing something...

I think this is the first 6th gen i7 thats obtained a better score than the 4790K systems out there right?

@AlexFielder is your system oveclocked at all?

Anyway you are "King of the Hill" at the moment! 🙂

Cheers

[AlexFielder]

My CPU is overclocked to ~4.5GHz, yes. (I guess you @Raider_71 haven't had any luck pulling the actual CPU speed at the time of the test from the available system information?)

It's odd that the tool still reports the main HDD as the Samsung 1TB storage drive when in fact it is an 850 Pro 500GB drive..?

I ran the new version of the tool one time, and got a result of 11.0 but then I ran it 3 times in the extensive mode and achieved this higher 11.9 score.

It may be something to do with the way Windows seems to increase the clock speed depending on what the system is doing; I don't fully understand the caching thing that even Windows 7 seems to do, but this higher result could be as a result of Windows 10 noticing that I was hammering this particular application a lot and it dedicated more processing power to it during the intensive testing? (This would need someone conversant with Windows 10 architecture to confirm)

I'll have it run the 10x tests later today and reply again with the results.

🙂

@Neil_Cross I would hope this machine would be top of the tree, given how much it cost me to build! 😉

[Neil_Cross]

@Raider_71 No I think I was missing my eyesight yesterday, I must have noticed that my modelling time was faster (4.92 against 4.98) and then when flicking between boxes, got confused over who's score I was looking at!

@AlexFielder My CPU was clocked at 4.8GHz for the test so I think your score highlights the benefits of the Skylake architecture, are you able to run the test at 1080p resolution or do you not have a monitor that can do that?

[-niels-]

Just wondering, but would putting inventor in "clean screen" mode during the test have any impact?

I wonder because it would at least set the screen resolution to more or less the same for everyone, the width of the browser bar wouldn't matter nor would the state of the ribbon...

[Raider_71]

Hi @-niels-

You have a very valid point there! I did some tests a while ago on my notebook and changing the size of the Browser had a fair impact on the graphics test.

I wonder if its not worth reporting on the size of the Inv graphics area as well? That way one would see the number of pixels your graphics card is actually having to work with.

[AlexFielder]

I shall certainly have a bash at that later on too.

I honestly expect the machine to perform exactly the same; Because of the single-threaded nature of Inventor, the system is just about breaking a sweat with this benchmark, but isn't being pushed to its limits like it would with perhaps a tool like 3DMark or Unigine Heaven real-time benchmark.

🙂

[Raider_71]

H @AlexFielder

I am still looking at a way to get the actual CPU speed. To be honest I have not focussed on that for a while but will start looking at that again soon .

The HDD reporting is according to the SATA port the drive has been plugged into. So yes on some systems the sequence is a bit odd.

Doing an extensive test should always yield better results as the time values (after the first run) will only update if the following test achieved a better score. So the time value will be the best out of n number of tests. I doubt that you will see the score jumping another IPI point when doing a 10 cycle extensive test though. Maybe 0.1 or 0.2 better but please let us know. 

As far as I know Windows does not control the CPU speed. Thats controled at a lower (hardware) level I would think. As soon as the CPU is loaded it should shoot up to maximum speed to try and deal with the task at hand. When no load the CPU speed throttles back to preserve power and lower heat generation.

[karthur1]

---

@Raider_71 wrote:

Hi @-niels-

You have a very valid point there! I did some tests a while ago on my notebook and changing the size of the Browser had a fair impact on the graphics test.

I wonder if its not worth reporting on the size of the Inv graphics area as well? That way one would see the number of pixels your graphics card is actually having to work with.

---

Just my two cents, but what if you just turned off the browser during the test.  That way everything is on a level field so to speak.

Kirk

[salariua]

Edit: got my menus and browsers mixed up.

Corrected:

It's not just the browser but the menu bar and status bar as well. I run Invetor with minimized panel all time. They take way to much space.

Using the "clean screen" command would be the answer I think.