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.
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.
Solved! Go to Solution.
Solved by Neil_Cross. Go to Solution.
Solved by Raider_71. Go to Solution.
Solved by Raider_71. Go to Solution.
Solved by Raider_71. Go to Solution.
@mo_mo wrote:
Hi,
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?
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".
Oh...
No. What's happened is the bench test has bugged out when it performed the visual test.
If you look at the scores in the graphics area for your test 2, I guess because it was pushed into the background the visual test bombed out after 0.05 seconds and recorded a reading of 7000Hz, which is absolutely impossible, but the test will still use that as a valid score when calculating the end result.
The test does have a message prompt saying to make sure Inventor is unobstructed and untouched throughout the test, I don't know for sure but I imagine DX11 freezes graphics windows when they're in the background to conserve system resource for power efficiency, because why have an application going visually at full capacity hammering the hardware if you can't see it.
I think the expression I'm looking for is...
remarkably average.
That's with Windows on High Performance mode, all cores locked and pathetically minimally overclocked at 3.9GHz (no boosting), DRAM at 2666MHz, SMT enabled as I can't turn it off on this motherboard yet.
In my manual 23 point stress test, Ryzen obviously destroys the ray tracing & cinebench tests, it competes quite admirably on a few other tests but in others it's being smashed by a $50 2 core Pentium G3258.
Yes, I think you're right.
No "hidden powers" after all. What a pity...
Sorry, that I posted the picture with the misleading score.
The idle temps aren't really a concern for me, they are a bit higher than Intel but not high enough to be a worry. I got a cheap-ish Corsair H60 for the Ryzen CPU and it peaks around 85C with an overclock.
Overclocking is pathetic. I have an old i5 in a cupboard that I managed to overclock by an extra 1.7GHz, this Ryzen CPU can't even take an extra 150MHz.
This was AMD's chance to take everything that people can do with Intel CPU's and improve on it, bring something new after half a dozen years of working on development, but instead it's just about competitive in one niché market and full of compromises and let downs in just about everything else.
I suspected the clock speed would have been the Achilles heel for it with Inventor, but with the entire Youtube/media tech space banging on about how it's such a great "all rounder workstation" platform, I just hoped they weren't all just spouting off words that they don't really understand, assuming because it's crap at most things then it must be good for that entire space. But basically it's only good for when an application can use 8 cores maxed out at 100% for long periods at a time, anything else and it's average at best.
A test I might do soon is under or over clock a load of my systems to 3.8GHz to match Ryzen, and see how it compares then.
@Neil_Cross may be interesting to see how it compare to an equivalent CPU with the same amount of cores. e.g. Intels' 6 core or 8 core chip, even and lower core Xeon chip?
I am disappointed at those results as I was potentially eyeing one of those up!
Don't get too hung up on the synthetic automated test results, it's not an accurate representation of real world performance. Although real world performance isn't brilliant, it isn't terrible either.
I've already put Ryzen through my 23 point stress test of real world workflows up against a 6 core Broadwell-EP Xeon, the E3-1650v4, I'm currently working on publishing that to my Youtube with the problem being I don't have enough hours in a day to get it finished in good time. Youtube takes more time to do than my day job so it's a constant fight with time!
@Neil_Cross What would you attribute the lower graphics test scores to? I know its not terribly low but I was looking back at the 4 tests I did with the GTX cards and they are all in the 7.xx range as opposed to the AMD card in the 11.xx (refering to the "category totals" box)
Hey Neil,
Interesting work you got going on there. I guess you could be the right person for this question.
Thru the entire post I see people showing off their machines and I have seen amazing benchmark that I envy but then will a double socket motherboard with 2 XEONs will make inventor fly?
Will there be any improvement running inventor under this specs (double socket motherboard and tons of RAM)? What's you opinion on that? I see people talking about Ryzen and the advantage of the cores but will two head brains be better than one?
I really enjoy your youtube videos very informative and keep up the great work man!!!! Thanks for helping out the community!
Thanks.
Many parts of Inventor are single threaded. So having a processor with many cores or two cpus will not gain you much. Inventor works best on high clocked cpus best. The i7 6700k and i7 7700k are the best cpus that I have seen for Inventor usage. Xeons will not perform as well due to their lower clock speed.
In short a single really fast cpu will beat anything slower with more cores.
Some of the newer Xeons are really impressive, I've got a HP Z440 here with a E3-1650v4 (6C) at 3.8-4.0Ghz and it's keeping up with my overclocked i7 at 4.6Ghz on a few of my stress tests.
Clock speed is only king scaled with architecture, I clocked an old CPU to 5.1Ghz and its slower at some single threaded tests than a newer CPU at 4.0Ghz.
I haven't managed to get my hands on a 6700k or a 7700k yet but they are indeed looking like the best so far, Ryzen was a bit a flop.
I wouldnt go as far to say that Ryzen was a flop, but if we were hoping for higher IPC's than intel, then yes Ryzen leaves much to be desired.
Sadly when it comes to this use case Ryzen is not even considered as an option. A $500 processor that keeps up neck and neck with Intel's $1000 enthusiast CPU is nothing to sneeze at. It's just too bad we cant use all 8 cores and 16 threads in Inventor. If that changes in a few years however.....
The first test I did the other day without changing any settings. The second test I ran this morning and I changed a number of settings, I deleted a number of idle processes, virus programs, and changed windows Performance Options to Performance.
Oddly enough it was a little quicker the first time though the graphics was a little quicker the second time (you would think) regardless they were pretty close.
My machine is four years old home built except for a new graphics card last month. Later this year I'll be investing in a new mobo, chip, RAM, and m.2 SSD and I'll retest to see how much improvement.
Sabertooth Z77 Mobo (Performance Mode)
i7 3770K 3.5 GHz
G.Skills RAM 1800 16Gb
Boot Drive: 256GB Samsung 840 Pro
Data Drive: WD Caviar Blue 750 Gb 7200 (though the program picked up one of my older HD's installed with a higher drive letter)
Asus GTX 1070 8 Gb Strix
To get a comparable result to the other test, you should have your screen set to 1920x1080. There was a previous post showing quite a significant drop in the score due to the increase in resolution.
Cheers,
Chris
A little off topic, we are K12 school district that uses HP workstations with K600 cards for Inventor and they seem to work fine
We would like students to be able to do Inventor and Revit work off campus via a virtual desktop
Two choices we are making right now for the 3 servers we are purchasing to host this environment
Which processor?
Xeon E5-2680v4 at 2.4GHz, 14 cores
Xeon E5-2690v4 at 2.6GHz 14 cores
In the server lineup the only Xeons that have clock rates above 3GHz have low core counts. I need to host about 140 VDI sessions and would rather not have to purchase additional servers to do so
Which graphics card?
nVidia Tesla M10
nVidia Tesla M60
The M60 is aimed at the professional graphics market. We have students learning Inventor so I am thinking the M10 might be enough
Any feedback is appreciated
So am I correct in saying that the AMD Ryzen looks faster on the benchmark scores that gamer's use, than the Intel equivalent but in reality it is slower than the intel as it has been optimised for the bench mark scores and not the really heavy number crunching that the real world deals with.
it like althanlon vs pentium chips all over again.