pullo

Benchmark 3.0

Recommended Posts

This is a continuation of Benchmark 2.0. I thought a new and clean slate would make things easier . 

The file for Benchmark 3.0 : 

Benchmark 3_0 for 2017.zip

 

1. So I propose Benchmark 3_0    . For starters the old file was a whopping  120 megabytes . This new file is  1.2 megabytes a 100 fold reduction in data transfer and you get the same stuff .  The earliest Mastercam I have is 2017 , so that will be the cutoff version .

2 NO WIGGLE ROOM HERE .  All the tolerenaces are set to 0.0001 for ALL the operation. Just open the file , go into the event log and with your right mouse button select clear log.

4.  The Mastercam event log is shown in the pic below . I suggest you open the log and with your Right Mouse Button (RMB )  "clear log" , for clarity.

 

5ac717e38d3ec_theeventlog.thumb.png.9ef91385a7bc0e6fdec1c6e827d0679f.png

 

 

Benchmark 3.0  Here it goes , the first log :

So the PC is off the shelf standard stuff, self assemled:

i7-7700K @ 4.2 GHZ  ,  

32 GB , SIMM 4 memory

Quadro K620  ( circa 250 Euro)

WINDOWS 10 v.1703

 

             Mastercam 2017 :  4min 46.5 sec

            Mastercam 2018 :   4min  44.8 sec

            Mastercam 2019 :   4min  12 .4sec

 

Conclusions :

1 .  As we had wiggle room before , my today's result shaved  a hefty  14 secs off my previous post (4 min26 ).

And I can't explain this , because I'd have to remember what I did to the file before etc. Now You open the file and let it rip .

DON'T MANIPULATE THE PARAMETERS OR WE ARE NOT GETTING RESULTS  WHICH CAN BE COMPARABLE.

2 . As one would expect there is improvement  in the newest version .  So no sense in comparing 2017 /  2018 to 2019 

3 I ran the test once more on 2019 and got :  4min 12.8 sec , so a 0.4 sec difference . So the result is repeatable now.

4 . I  had a virtual drive installed in my previous  PC and I had all my Mcam temp files created on it, but as far as I remember , there was no significant time reduction 

when using it. 

 

I'm sorry I had to "hijack"  the Benchmark 2.0 thread , but I saw so much inconsistency in the method , I felt putting this thing back on track was the right thing to do. And thanx previous post for wanting results for different  Mcams .  Mastercam 2019 is the way to go in my opinion :). 

My next step here is I bought a water cooling set and I'll try to bump up my  Gigahertzes up to 4.7  GHZ.  I predict my time will decrease in a linear fashion  at ratio of 4.2/4.7 . I'll post the results when I get this done.

 

 

Gracjan

 

2017

2017.thumb.png.2b39d805c1548e9199db8553385cb4ad.png

 

 

2018

 

2018.thumb.png.acd6642ee67702d9b6f308077dd3e9b4.png

 

2019

 

2019.thumb.png.dedfcf536e6d1ed1155343963c41d459.png

Edited by Michael Gingerich (IHS)
Benchmark 3_0 for 2017.zip link updated
  • Thanks 1
  • Like 4

Share this post


Link to post
Share on other sites

I left out the self evident stuff ( to me) out ,  but just in case  I'll add them here so that one does not have look somewhere else.:

1.Download the Benchmark 3.0 file and open it with Mastercam.

2. The best and only way (as toolpaths will be calculated  parrallel where possible)  to  determine the  speed calculations is to look at the times reported by Mastercam  event log.

This is a separate program entity that starts up in the background . It can be accessed via the "Show hidden buttons" situated in the bottom right  corner. As there might be lots of 

report lines already there,  one can get rid of all the  report lines by clicking  the Right Mouse Button (RMB)  and choosing the "Clear  Log"  option.

3 . Choose all the operations  in the ops manager and click "Regen all selected operations". And now we wait for from  3 to 25 minutes for all the ops to be recalculated.

4. Now add all the times together.  I usually covert all the times into seconds in me head and then divide the end result by 60 . That will give the minutes. Subtract the whole minutes times 60 from the total second result and you are left with the seconds. 

5.   The screen grabs at the end of the previous  post  are from the event logger. The logger logs everything Mastercamy  , so you can have different  sessions  open  2017 , 2018 etc and every movement in Mastercam will be recorded.

Gracjan

 

 

Share this post


Link to post
Share on other sites

My  previous PC :

i7-4790K @4.00 GHZ 

32 GB SIMM3 memory 

WIN 10  Build 10240 ( a very old version from 2015 )

Mastercam 2019  : 5min 58 sec 

 

Conclusion :  The processor gigahertz ratio would have warranted only a 5% decrease  in  calculation time.  It's the SIMM3  to SIMM4  speed decrease that gives me a 19%  overall speed reduction when compared  to  the 4.2GHZ machine  . It could be that the processor is faster overall...

 

Gracjan

 

 

 

 

 

 

 

Share this post


Link to post
Share on other sites

Are you setting the mutil thread manager to use all available cores?

I downloaded your new file this morning

Set x19 to use all available cores and got a run nearly a minute slower than yesterday  LOL

What tolerance did you set the stock model at.

I was running at .001 yesterday

You also have to remember to clean out the regen folder between runs

  • Haha 1

Share this post


Link to post
Share on other sites

This is with 12 tabs open in Chrome and also streaming the Masters :P

 

Capture.JPG

Share this post


Link to post
Share on other sites

Dang... ran it... still 8 minutes... still need a new rig.

Jeff... if you can... PM me your info on where you got your computer and its specs.

My boss told me he'd spring for a new one... might as well get a beast like yours.

Share this post


Link to post
Share on other sites
7 minutes ago, Reko said:

Dang... ran it... still 8 minutes... still need a new rig.

Jeff... if you can... PM me your info on where you got your computer and its specs.

My boss told me he'd spring for a new one... might as well get a beast like yours.

PM sent.

  • Like 1

Share this post


Link to post
Share on other sites
2 hours ago, gcode said:

Are you setting the mutil thread manager to use all available cores?

How do you do that?

 

Share this post


Link to post
Share on other sites
10 minutes ago, So not a Guru said:

How do you do that?

 

View/Multi Threading Manager/ click the !

OR

Alt+M

 

Mine is set to 4 threads, probably could have set it to 8

  • Thanks 2

Share this post


Link to post
Share on other sites
5 hours ago, pullo said:

Now add all the times together.  I usually covert all the times into seconds in me head and then divide the end result by 60 . That will give the minutes. Subtract the whole minutes times 60 from the total second result and you are left with the seconds. 

Won't subtracting the start time from the finish time, in the event log, give the same result?

Share this post


Link to post
Share on other sites
12 minutes ago, So not a Guru said:

Won't subtracting the start time from the finish time, in the event log, give the same result?

It seems to give a slightly different result, but I would argue that's the one that matters.  By my math 4:27 from Gracjan's 2019 event log.

Share this post


Link to post
Share on other sites

Not bad for a Xeon powered laptop.

image.png.0083804a168d7adace41e3a9e06203ed.png

 

2018

image.png.000748cf440ad42f79609d746c488216.png

Share this post


Link to post
Share on other sites

I will try overclocking when I get brave enough, but I'm ok with this.

 

benchmark 3.PNG

Share this post


Link to post
Share on other sites
17 hours ago, gcode said:

Are you setting the mutil thread manager to use all available cores?

I downloaded your new file this morning

Set x19 to use all available cores and got a run nearly a minute slower than yesterday  LOL

What tolerance did you set the stock model at.

I was running at .001 yesterday

You also have to remember to clean out the regen folder between runs

I think I was in a pretty good position  to start off this new thread  ,  as the  hardware that matters  here is  the  fastest possible on the market (processor+memory) priced reasonably.  Also  nothing whatsoever is souped up  so far . OK I've read  about some new  processors, but I'm not putting  money into a processor whose price does not follow a linear price-speed increase as to where I am now ( and the line is at least at an angle of 45 degs :) ) . I already got burnt with the whole XEON  shebang...

The multithreading is  set from  File-Configuration-Toolpaths and Enable Multithreading  on .

gcode , I noticed you were running the test at a tolerance ten times  more loose  then the rest of us .  That is why  you got a low 4 mins.  I  am not going to go there because  the moment we start tweaking the test file  ,  comparability  goes out of  the window....

Some of you don't  add up your times.... Me I just look at the last line of the times , if it's  smaller than 2 minutes for the old timers ( 2017 and 2018 )  and  if it's smaller that 1.5 mins for the millennials version :) (2019 )  , that will get my attention .

Orvie ,  you were quite modest about your rig , that Intel I9  7980XE  processor is a whopping 2100 Euros , but  you only got 8 seconds (4min 36 sec vs my 4min 4 4 secs) off my 2018  result :) .  That is what I call a non-linear speed increase :) 

gcode ,  my  Regen  directory was  completely empty .  I never save the  test file , just open  it  from the downloaded  .zip file  again and again (I uploaded it myself I know , but mistakes can happen )

and then the methodology is compromised). 

Speed of course usually comes at some  price .   

My machine was about 1500 E ,

400E for the processor (4.2 GHZ) ,

500E for the memory (SIMM4),

250 for the  Graphics card

and the rest for a new  box , 500 GB  SSD  , etc...

I looked up my processor and it's price has come down to 300E over a period of a few months so it's 

BTW ,  I don't consider myself a geek , I have a life  and a wonderful family .  I'm just curious :)

Gracjan

 

Share this post


Link to post
Share on other sites

Correction ,  the overall multithreading is set  from  the  configuration : 

 

 

 

5ac870faf16db_mtconf.thumb.png.9b97770d515ad4a9b1e5015484d7f943.png

 

but I personally totally missed the wise words of Jeff  first time around ,  so I'll spell that  out ,  in order that one session of Mastercam  does not hog all the threads , 

VIEW / Multi-Threading  AND THEN THE EXCLAMATION MARK :

threads.thumb.png.109174c96ab617f1b72c3d3fee6208fb.png

 

Gracjan

 

 

 

Share this post


Link to post
Share on other sites

I ran this twice this morning

Once with multithreading set for 4 cores, once for 8 cores

4 core time was 5:11 

8 core time was 5:07

calculating the time is easy 

save the event log as a .cvs file, open it in Excel and let Excel do all the work

see attached file

gcode_Benchmark 3.0 X19_PB4.xlsx

Share this post


Link to post
Share on other sites

I'm  guessing gcode you have  SIMM3 memory , because you should be getting better times wit @4.5GHZ

Gracjan

Share this post


Link to post
Share on other sites

2018:

Capture.thumb.JPG.91000242d840679c5f5c1eb97b11df5b.JPG

cpu.JPG.45f2b9dd40037e5378b0f030d1139106.JPG

HDD is a Samsung 960PRO

Ram is @ 4000mhz

 

 

  • Like 2

Share this post


Link to post
Share on other sites

what was your starting  frequency here ?

Gracjan

Share this post


Link to post
Share on other sites
6 hours ago, pullo said:

what was your starting  frequency here ?

Gracjan

The motherboard is an Asus Gryphon Z97 which is about 4 year old technology

Processor is a  i7-4790  4.0Ghz overclocked to 4.5 Ghz

The mother board is maxed out at 32 gig  of DDR3-1600 ram

There is something that confuses me though.

The specs on the webpage show max ram speed at 1866

Quote

4 x DIMM, Max. 32GB, DDR3 1866*/1600/1333/1066 MHz Non-ECC, Un-buffered Memory *
Dual Channel Memory Architecture
Supports Intel® Extreme Memory Profile (XMP)
* Hyper DIMM support is subject to the physical characteristics of individual CPUs. 
** Refer to www.asus.com for the Memory QVL (Qualified Vendors Lists).

yet the Qualified Vendor list for this board lists ram up to DDR3 3200

GRYPHON_Z97_Series_memory_QVL.pdf

Share this post


Link to post
Share on other sites
20 hours ago, gcode said:

 

There is something that confuses me though.

The specs on the webpage show max ram speed at 1866

yet the Qualified Vendor list for this board lists ram up to DDR3 3200

GRYPHON_Z97_Series_memory_QVL.pdf

You have to go into your BIOS to overclock the memory.  

Look for SMP profile of something of that nature and change your RAM speed to what it's rated for.

Share this post


Link to post
Share on other sites

Thanks for reviving the technique and getting it back out there.  I have been thinking about building a tower, just haven't had time nor the money for it.  The more people that throw their specs out there now, the less time it will take to put a cart or parts together.

5:53 in 2017  running with 4 cores

6:21 in 2018 running with 8 cores

Hmmmm, not sure why 2018 was so much slower.  Anyway, not bad for a 4 year old laptop.  At the moment, my normal regen times don't get much worse than 2 or 3 minutes, and that is on my multiaxis paths.  I usually rough and finishing in one shot cutting wood, and the steopovers and tolerances are relatively loose at between .0005" -  .002" for the tolerance, and .030" - .125" for the steopover.  I don't use stock model much because with one of the tools I use it is guaranteed to crash (logged bug), that seems to be the slowest regen item usually.  So maybe a new rig can be postponed for a while yet again.

How many of you guys are running a page file?  Mine has been off for a long while now.

Share this post


Link to post
Share on other sites

Hi Husker . It's Gigahertzes in the processor and the fastest memory  that will get you inner peace.  I once again tried a virtual disk and directed all the temp files there. Niente on that lead :)

Gracjan

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

Join us!

eMastercam - your online source for all things Mastercam.

Together, we are the strongest Mastercam community on the web with over 56,000 members, and our online store offers a wide selection of training materials for all applications and skill levels.

Follow us