Benchmark

[Jayson Kramer]

Dave we need computer specs sir.

[djstedman]

Dave,

 

Though im sure your correct about the benchmark file being able to be tweaked to make it better for efficiency and for real world applications, the point originally I beleive was just the opposite.. to find something that beat the heck outta the system for benchmarking against other systems because the harder it makes the system work the better it is at pointing out which systems are actually performing better.

 

After all .. pretty much any system will regen a simple contour within 10ms of each other but that doesn't mean the two machines are equivalent.

[Scott Lundin Shopware Inc.]

Using the new bench mark file that was posted on Jan 12, 2012 with the computer specs listed below:

 

X6MU3 = 7min 55 sec

X7PC1 = 4 min 53 sec

 

X6 with all products(dealer sim)

Intel I7-2600K

Asus P8Z68-V/GEN3

16GB Gskill Ripjaws F3-12800CL9D-8GBXL

Quadro 2000 video card

standard 7200 RPM hard drive

2 x 500GB hard drives in raid_1 for data

[Zoffen]

Just got my workstation put back together after a weekend of overclocking!

 

 

4 monitors = bare minimum

 

I have a more stable overclock now, ill run the benchmark again and see what i get!

 

I'm wondering if we should add some multiaxis paths to the benchmark.

 

I know when i crunch multiaxis paths i get full core usage for alot of the collision checking stuff but when i crunch high speed paths its never full bore core utilization like multiaxis seems to be.

 

just throwin it out there.....

[Zoffen]

Ok so i ran the benchmark again.

 

this time running at 4.6ghz & 2133mhz ram

 

This time it took 7:10

 

 

I even regened the operations at .005 tol then did the benchmark at .0001 tol.

[Thee Rickster ™]

Just got a new p.c. at work today up and running

 

i7-3770K @ 3.5GHz

Win 7x64

32 gig ram

250 gig ssd

Quadro K4000 video

 

43 seconds

[gcode]

Your going to like that rig

[mkd]

Just got my workstation put back together after a weekend of overclocking!

 

 

4 monitors = bare minimum

 

I have a more stable overclock now, ill run the benchmark again and see what i get!

 

I'm wondering if we should add some multiaxis paths to the benchmark.

 

I know when i crunch multiaxis paths i get full core usage for alot of the collision checking stuff but when i crunch high speed paths its never full bore core utilization like multiaxis seems to be.

 

just throwin it out there.....

next time you take a picture of your desk, remember to put away the pot-pipe!

 

 

made u look

[Zoffen]

next time you take a picture of your desk, remember to put away the pot-pipe!

 

 

made u look

 

lol.....

 

or other things you forget to put away....

[Radical1]

Sorry to be a PITA but I don't run High Speed Toolpaths, I was going to benchmark this past weekend not sure if I understand the instructions properly. Am I supposed to change the ARC filter in the parameters of each of the last three toolpaths to 0.0001 then regenerate them and time them?

[Newbeeee™]

Well this was on one of the previous PC's running the original benchmark file

X5mu1

Intel i7 950 4core 3ghz

8meg cache

10GB DDR3 1333mhx

GTX480 (cheapskate - thought I'd give it a go)

Windowze 7 64bit

1:04

 

On new PC

Intel i7-3930 3.2ghz 6 core

12meg cache

16 GB memory

GTX680 (gotta lurve saving a bit of money over those quaddro cards )

Windowze 7 64bit

45.3 secs

 

As soon as SP2 comes out, I'm gunna go 7 - Woohoo!

[Rotary Ninja]

01:08 Time to Process

 

PC is:

Dell OptiPlex 790

intel core i5 2400 @ 3.10 GHz

4GB Memory

Windows 7 Pro 64 bit

Nvidia Quadro 2000D

 

Getting a new PC today so we will see how it goes.

[Jayson Kramer]

Newbee and Rotary you must be running the original file and not the updated with Stock model correct?

[Newbeeee™]

Yes Jay - I'm so last week

[Jayson Kramer]

Ok newbeeee we won't tell if you don't. So anyone else reading this you are not really reading this .

[Matthew Hajicek - Singularity]

GTX680 (gotta lurve saving a bit of money over those quaddro cards )

 

I was having hella glitches with my Quadro 2000. I switched back to the GTX 560, the worst of the glitches went away, and I haven't noticed any performance hit.

[Newbeeee™]

I was having hella glitches with my Quadro 2000. I switched back to the GTX 560, the worst of the glitches went away, and I haven't noticed any performance hit.

Matt - you tight arse cheap skate ....welcome to the club

[Rotary Ninja]

Ok, I ran the new file on my 3 year old AMD PC. Here are the specs...

 

Windows 7 Ultimate SP1

Gigabyte GA-MA785GMT-UD2H Motherboard

AMD Phenom II X4 840 @ 3.2GHz

16 GB Corsair Vengence running at 1333 MHz

120GB Corsair Force GT SSD Boot Drive

Quadro 2000D Video Card

25" HP 2509B Monitor

 

Mastercam X6 Home Learning Edition

 

It regenerated in about 18 minutes...

[Xico]

Hi!

 

I don't understand why the verify is so slow in some simulations with small distance between paths. I try to use a new notebook (a good one) but the the results are a little bit better than my notebook with 7 years...

What is need to accelerate the verify? The notebook that I tried have 32gb of ram, a i7 3560, a GTX 670mx card and a ssd.

What can be better to the verify goes faster? I'm using the mastercam X6.

 

PS: The turbo mode don't allow a model in the stock only a box and the precision is less in the comparison with the stl. I use best quality with the 100 in the moves/refresh.

 

Please help me to understand this.

[Big Fish]

using the "new" x6 file

 

X6 MU3 = 00:09:06

X7 SP1 = 00:06:21

 

a little surprised. got a cut direction compatibility error in x7 (which it "auto updated").

[gcode]

using the "new" x6 file

 

X6 MU3 = 00:09:06

X7 SP1 = 00:06:21

 

a little surprised. got a cut direction compatibility error in x7 (which it "auto updated").

see page 52 and 53 of the What's New Files for an explanation of the compatibility error

 

C:\Program Files\mcamx7\documentation\WhatsNew.pdf

[Big Fish]

see page 52 and 53 of the What's New Files for an explanation of the compatibility error

 

C:\Program Files\mcamx7\documentation\WhatsNew.pdf

 

yeah i've run into it before and probably looked it up the first time around. I was just noting that the x7 test file had been altered to x7 compatibility. I was surprised by how much more quickly it went in x7.

[Guest]

any SSD vs HDD comparisons on X7?

[Guest]

That would be interesting.

 

I did one in X5 and the time was significant. Justified the price in a hurry.

[Matthew Hajicek - Singularity]

Could one of you guys with six or more cores tell us if verify will use all of them? I notice it will peg my four cores to 100%, so that's a possible bottleneck.