Benchmark

[gcode]

If you installed the inch sample files, you will find it there

[jeff]

50.403 seconds

 

i7 2600k (stock clock)

GTX580Ti

120GB SSD

8GB RAM

24" Asus HDMI monitor

 

$1928 out the door with 3 yr warranty.

Compared to the uber expensive pc's that are being pushed on here, I'd say it's just as fast if not faster.

 

 

[Brian B 74]

32.887s

 

MCX 6 - 64 bit

i7-2600 @ 3.40GHz

8GB RAM

nVidia Quadro 2000D

Win 7 Pro - 64 Bit

[jeff]

Attempt #2 was 40.103 seconds.

 

Solidboxxxx? BAHAHAH!!! BAAAHAAAAAHAAAHAA!!!

[Thee Rickster ™]

I upgraded to win7 64 and doubled the ram

 

Core 2 duo 3.16ghz

8 gb ram

win 7 ultimate 64 bit

gtx 250

 

2.06

 

 

 

same box with X6

 

1.29

[neurosis]

My current box with results. I want to compare when I get the new pc.

 

 

So,

 

Dell Vostro 400

Windows xp pro sp3

4 gigs ram

Intel Core 2 Quad Q6600 @ 2400 Mhz

Quadro FX1500

Mastercam X6 32bit

 

2 mintues 13 seconds.

 

 

I think that we ought to change these tool paths to some of the newer paths and create a new bench mark. Maybe use optirough, optirest, horizontal, and hybrid with .0001 tolerance. Just to make it interesting.

[Joe788]

My current box with results. I want to compare when I get the new pc.

 

 

So,

 

Dell Vostro 400

Windows xp pro sp3

4 gigs ram

Intel Core 2 Quad Q6600 @ 2400 Mhz

Quadro FX1500

Mastercam X6 32bit

 

2 mintues 13 seconds.

 

 

I think that we ought to change these tool paths to some of the newer paths and create a new bench mark. Maybe use optirough, optirest, horizontal, and hybrid with .0001 tolerance. Just to make it interesting.

 

2 minutes! Ouch! You're gonna be happy when you get your new box built. I spent $1300 on mine starting with a ZT Systems Window 7 64 bit box from Costco, installing an SSD as the boot drive, and a POS FX580 graphics card, and I'm at 40 seconds in X6.

[Thee Starry™]

39.6 seconds

 

Dell

Precision T7500

Intel Xeon X5690 @ 3.47

12Gb Ram

Windows 7 64 bit OS

Nvidia Quadro 4000

[neurosis]

Ok guys. I created a bench mark with updated tool paths for you guys with the bad a$$ pc's. It is an X6 file using optirough, stock model defined by stock + optirough, optirest, hss horizontal (same as before) and hybrid.

 

Here is the file

 

Its on the ftp under unspecified uploads since there are no folders for x6 yet.

 

open the file and dirty the operations. Change the filter to .005 to dirty, and then back to .0001 to run.

 

As we all know, just dirtying the optirough will automatically dirty the stock model.

 

Lets see now fast you calculate this sucker.

 

Then if you are really brave, tell me how long it takes to simulate it with your quality turned all the way up.

[Matthew Hajicek - Singularity]

Looking at an upgrade. Considering these components:

 

i7-2600 http://www.newegg.com/Product/Product.aspx?Item=N82E16819115071

 

120GB PCIE SSD http://www.newegg.com/Product/Product.aspx?Item=N82E16820227745

 

16GB DDR3 2133 http://www.newegg.com/Product/Product.aspx?Item=N82E16820231503

 

Which mobo should I glue them together with? Is it worth messing with the PCIE SSD, or should I go with the standard SATA type?

[Matthew Hajicek - Singularity]

Ok guys. I created a bench mark with updated tool paths for you guys with the bad a$$ pc's

 

Downloading...

[neurosis]

LMAO! 16:40 and I can barely rotate it when the stock model is selected to watch the rest-rough run.

 

I am also noticing that while opti/dynamic has support for bullmills now, I am not so sure that it supports ballnose very well. Ive seen a couple of back feed moves that shot through cusps.

 

Matt,

 

Forgive me if I am mistaken, but I dont think that ram will work in a 1155 board. You will need an X79 board (2011 socket) I believe.

 

I thought about stepping up to the 2011 board and getting a Intel Core i7-3930K @ 3.20GHz but the boards are too new.

 

I'd make sure that you know what you are getting before you get it.

[Matthew Hajicek - Singularity]

13:48, though I noticed Op 3 generated twice. Once before the stock model and once after.

 

I can rotate the stock model fine though.

[gcode]

About 10 minutes ..

see the attached Event log screen shot

what I can't figure out is how it regenned all the ops and the stock model last

doesn't it need the stock model to regen OP#3??

 

edit: no problem rotating the stock model..

[Colin Gilchrist]

Tom,

 

Looks like it finished regenerating #2 (stock model), then regenerated #3. Unless I'm missing something...

[gcode]

yeah .. I was looking at it wrong...

 

I should note I ran that time on my work machine , not the machine in my siggy,,

 

This is a Dell T5500/Solidbox with a

X5670 Xeon @ 2.93 GHz

12 gig ram

Quadro FX-4800

dual 10k WD hard drives in RAID 1

 

my home machine should be a little quicker

[neurosis]

Yes, time to replace my fx1500. I think that it may be getting a bit outdated. I have a quadro 2000 on the way. Should be to our shop before the computer is done.

[GF8er]

Little over 6 mins

 

intel i7 2600k overclocked to 4.4ghz

8gb ddr3 ram

120gb ssd

nvidia quadro 600

win 7 pro 64 bit

 

[Izzzy]

Running in X5mu1 the original benchmark file.

 

2 Mins

 

Dell Precision T5400

Windows XP pro x64 32 GB of RAM

8 Core Xeon X5450 @ 3.00GHz

Quadro FX-3800

[Matthew Hajicek - Singularity]

Looks like it finished regenerating #2 (stock model), then regenerated #3. Unless I'm missing something...

 

Yep, it looks like it's always regenerating op 3 twice, once before the stock model and once after.

[Brian B 74]

Might want to consider dirtying the toolpaths for the download. My file size with dirty toolpaths is 1.09MB.

 

Just my $.02

[Matthew Hajicek - Singularity]

That would help avoid differences in regen conditions too, if it's all set to go.

[David Conigliaro CNC Software Inc.]

LMAO! 16:40 and I can barely rotate it when the stock model is selected to watch the rest-rough run.

 

I am also noticing that while opti/dynamic has support for bullmills now, I am not so sure that it supports ballnose very well. Ive seen a couple of back feed moves that shot through cusps.

 

Matt,

 

Forgive me if I am mistaken, but I dont think that ram will work in a 1155 board. You will need an X79 board (2011 socket) I believe.

 

I thought about stepping up to the 2011 board and getting a Intel Core i7-3930K @ 3.20GHz but the boards are too new.

 

I'd make sure that you know what you are getting before you get it.

 

 

Bullnose are treated like bullnose, flat are treated like flat, ball are treated like flat. To treat ball like true ball or bull is not going to yield very effective toolpaths. Dynamic motion is designed to run bull or flat specifically, it's a full depth roughing toolpath. If you want to see what you ball would yield, define it as a bullnose. So a 1/2inch ball could be defined as 1/2 bull with .245 rads. But your max stepover will be severely limited due to the small flat of the bull.....

[David Conigliaro CNC Software Inc.]

Neurosis,

We have thought about ball tools and chose to support them like flat endmills. You're basically expecting a ball nose to remove cusps after roughing or to avoid leaving cusps on flats, that's a lot of unnecessary motion due to a tooling choice. Furthermore, we cannot predict or detect where the cusps are to ensure micro-lifts avoid them. The best advice I can give you is to calculate the cusp height for your cut condition, which is simple since your on a flat floor, and enter an appropriate micro-lift height. Cheers.

[neurosis]

Thanks Dave, that makes sense.