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.

[Anonymous]

---

@Neil_Cross wrote:

Good score that @Anonymous

---

Thanks, Neil. We outsource the IT and I think the consultants did a pretty good job of building, choosing parts for the PCs. They're pretty minimal in terms of what's running in the background.

[karthur1]

@Anonymous.... that would be  IPI=9.09,  That's the second highest I have seen.  That is good!

[Neil_Cross]

Third highest now 

[blair]

I wonder if I pull some of the liquid Oxygen from our 1500gal storage tank and super cool my system and overclock the crap out of it. What's the worst that could happen, need to put in a new MOB, CPU and possibly a new Titan graphics card. 

[hosford]

Blair

Much better to use liquid nitrogen, thinking liquid oxygen might have some negative results.

see this for some of the fireworks https://www.youtube.com/watch?v=UjPxDOEdsX8

[blair]

Everything burns better with 100% Oxygen, even stuff that shouldn't burn.

[Raider_71]

Hi guys,

Ok here is the latest incarnation of InventoBench. Version 1.3.1 C

The dropbox link is still the same as per the first post but there is a new A360 link

This version has the following new features:

1. The introduction of the “Inventor PC Index” or IPI. This number basically looks at an overall performance value and consists of two parts. The value before the hyphen “-“ indicates the test version. In this case it’s a “1” as this version of InventorBench is still using the initial testing procedures since the release of the tool. As soon as there is a change in the testing procedures then this number will increment to 2 to indicate that it’s a newer IPI number.

The second number is the performance indicator and the higher the number the better overall performance your PC has with Inventor. This number is derived from all the results accumulated except for the “Warm Start” value. We have found that there are too many factors that could affect this number and this would skew the overall number too much. The Vault Add-in for instance could effect this number by a good number of seconds which in turn would have effected your IPI value so we decided to remove it for now.

So the HDD, Modelling, Drawing and Graphics card test results are used to determine the second part of the IPI number.

1. The ability to run multiple test cycles. The idea behind this is to determine the best score your PC can achieve with its current configuration. So after n cycles the results you see will be the best your computer could achieve in each category out of all the cycles. With this option you should be able to get a much more consistent results for a system.

As always please keep the suggestions coming as this helps to decide what to focus on next. There are a few requests I sam still looking at building in but I will keep posting here for now with new builds.

Cheers

Pieter

[salariua]

Playing with oxygen is very dangerous. We had and oxygen plant mainly to assemble stuff (press fitting) and I remember a guy trowing a greased nut into the barrel when there was just a cup left and it thrown that barrel way up in the air. Now that's stupid!

Nitrogen cooling might give us the best inventor workstation; its CPU that we're after. Anyone still remembers the old TomsHardware project?

Pentium 4, 3.2 GHz OC to 5.25 GHz

[karthur1]

Never heard of anyone doing shrink fits with liquid Oxygen.  I guess you could, but that could be very dangerous.  We use Nitrogen pretty often here since it is inert and does a good job.  They both have the same boiling point of around -300°f.

Kirk

[Anonymous]

---

@Neil_Cross wrote:

Third highest now 

---

My settings were at Quality. I'm sure at performance, I could top that.

[Neil_Cross]

---

@Anonymous wrote:

---

@Neil_Cross wrote:

Third highest now 

---

My settings were at Quality. I'm sure at performance, I could top that.

---

Give it a try, but the only thing the Quality option does is apply a minimal amount of anti-aliasing to the visual display of the model.  Between Quality and Performance you'll only see a handful of FPS increase on the visual test and no change to the CPU and drawing tests.  

[-niels-]

Just wanted to throw my results in again, since the new version has an easy comparison number.

I switched to performance and like Neil says, only a slight increase in FPS.

I did a 3_cycle extensive test for these results, on Inventor 2016 SP1 (actually R3, but it doesn't say that in the about 😉 )

Which brings me to my suggestion @Raider_71 , since everyone is using this wonderful tool now and comparing it here:

Shouldn't there be a mention of which inventor build the test is run on?

Not sure how much is matters, but if they've changed the render engine for 2016 wouldn't it be possible this might have some effect on the graphic tests compared to other releases?

Oh and @Neil_Cross, i just realized... i ran (both) my test(s) while connected to wifi, how much can network traffic mess with any of these tests?

[Neil_Cross]

The version of Inventor is mentioned in the test results, on the left hand side, version and service pack level!

Being connected to wi-fi shouldn't make even the slightest bit of difference to the results, the test makes zero use of any network or internet resource so this shouldn't be a factor whatsoever.

Having said that... if your styles library or some project paths are network based and you have a laggy weak connection over wi-fi, this could effect some of the results if Inventor is chugging away trying to resolve paths, especially with the warm start time.  @Raider_71 may have factored this into the test though by switching to the default project file for the duration of the test.

[-niels-]

---

@Neil_Cross wrote:

The version of Inventor is mentioned in the test results, on the left hand side, version and service pack level!

---

"go sits in a corner ashamed of not seeing that"

[Raider_71]

Hi @-niels-

As Neil said, the Inventor version and SP level is reported on the info screen unless you are referring to the minor build numbers? ...which I dont believe will make a big difference but maybe worth testing 🙂

@Neil_Cross you are correct in saying that I am avoiding the affect of style loading, templates and other potentially shared network resources and that the time measurement is specifically done so that the result pertains to the task, as much as possible.

Please keep the ideas coming. 

Pieter

[Anonymous]

So, I grabbed a R9 380 from an idle computer and compared it to my K4200. Also was able to OC to 4.1ghz.  I don't know if I'm going to go higher.

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

New bench:

[Neil_Cross]

I think it's more about the CPU, someone has posted a result with a Quadro K6000 installed and his visual scores where around 120fps max.  Both @Anonymous and myself have the i7-4790K which is so far proving to be one of the best all round CPU's for Inventor, 4.4GHz clock speed is nailing it.  My last post was the 4790K cranked up to 4.8GHz.

@Raider_71 I still think there's something iffy with the FPS reporting.  My realtime FPS display is reporting almost double during the test what the bench tool is recording once the test completes.  At what point does the bench tool snapshot the FPS result? 

[blair]

It really looks like you want to invest your money if the fastest CPU and save money on the graphics card. All the heavy lifting is done at the CPU level and as long as you have adequate screen refresh of 32 fps of faster for smooth panning/rotation you are fine.

[Anonymous]

So I did get it to safely OC to 4.5.  Score went from 9.1 to 9.5.

I guess I should actually do some work today.  I can play with this for a while...

[Anonymous]

@niel_cross what going on with this modeling time then.