Jump to content

Welcome to eMastercam

Register now to participate in the forums, access the download area, buy Mastercam training materials, post processors and more. This message will be removed once you have signed in.

Use your display name or email address to sign in:

Benchmark


Roger Peterson
 Share

Recommended Posts

quote:

He says this (mcam) won't take advantage of a Tesla cuda-based machine. Too bad.

What have I been saying this whole time! Abandon the cpu and move it to GPU. CUDA or OpenCL!!! But what do I know about computers :rollseyes:

 

cnc software, if you want to make mastercam fast, look into CUDA based threading or Open CL based. Then we could take advantage of tesla/fermi as computing cards or just use the ones we have in our comps!

Link to comment
Share on other sites

Hi all!

 

I found an easy way to significantly improve calculation times on i7. My original time was around 1:30 with my i7 model 920, but I was able to get 1:13 by changing the default BCLK (aka. base frequency) from 133 MHz to 200 MHz. As I understand it, this one setting affects memory, CPU and BUS speeds. Basically my processor went from 2,7 GHz to 4,2 GHz.

 

304.png

 

These setting are machine specific, so you need to google or test the right setting. And of course you can get even more by changing voltages etc, but that is more complicated and it easier to break something. When I tried with custom voltages etc, it toke over two minutes to complete.

 

We tried this on Reijo’s i7, which is model 860, and there 180 MHz was too much. We settled for 150 MHz, even though it probably could take more. The original time was 1:48 and with the change, it was 1:32.

 

Some motherboards, like mine X58M, have this ability to test the highest setting automatically. In my case, it didn’t work, but it should be the safest way, so you should at least try it.

 

Here’s my hardware:

Intel i7 920 over clocked to 4.2 GHz

MSi X58M board

6144 MB of DDR3 memory

NVIDIA Quadro FX 1800

Windows 7 64-bit

 

I also changed the Affinity settings from the task manager, since for some reason, the default for Mastercam in my system is only one core. With only one selected, it calculates 50% slower, since it using, one core. So by clicking the Mastercam.exe, from the process-menu, with my right mouse button, I can get to the affinity settings. There I can choose that Mastercam.exe can use all cores.

 

Edit:

I would also like to add, that the multi-threading didn't make any differense to the calculations. The time was the same.

Link to comment
Share on other sites

As I understand it, my i7 doesn't have a temperature meter. The temperature is calculated from voltages and some other integers. When the calculated value reaches 100, the processor starts to drop the multiplier to slow down the processor and keep the calculated value under 100.

 

In my tests, this really hapens. After several minutes off full load, the temperature integer value get's to 100, and the multiplier drops.

 

But a better cooler could not hurt.

Link to comment
Share on other sites

My T3500 came in today, and the benchmark was the first thing on my to do list.

 

I spoke with Mike Debutts earlier this week about a Solidbox system, but I couldn't wait 3 weeks for a new box.

 

My laptop finally bought the farm, and as you can see, I have a stack of new jobs and rev changes on my desk. There are 2 more stacks that go with that one.

 

I found this on the Dell refurb site for $2500 with the monitors included.

 

Dell Precision T3500

Xeon W3570 3.2Ghz 8M L3 6.4Gts

4 Gb ECC DDR3 @ 1333 Mhz

768 Mb Nvidia Quadro FX1800

300 Gb 10,000 rpm Velociraptor 16Mb Cache

Windows 7 Professional 64 bit

Dual P2210 22" Widescreen Monitors

SpacePilot Pro

 

So far I love it.

 

I'll bump the memory to 12 Gigs and add a second Velociraptor in a couple months and benchmark it again.

 

 

Meh new box:

 

photo.jpg

 

 

Respectable time for a $2500 setup.

 

Benchmark.jpg

Link to comment
Share on other sites
  • 2 weeks later...

I just installed my second Velociraptor today and set up raid 0 through the Matrix onboard raid control and only see a second or two gain in the benchmark.

 

Is this all I should expect? Should I format and reinstall? headscratch.gif

 

Everything says that the array is working properly, but in windows disk management, it is still recognized as a simple volume, not a dynamic or striped volume. I'm guessing that's because the raid array is controlled by the bios.

Link to comment
Share on other sites

5 min. 38 sec.

IBM Intellistation M Pro (desktop)

Pent. 4 @ 3.2 GHz

2GB Ram @ 800 MHz

Windows XP Pro

 

Total tolerance all 3 ops set to .0001

 

I was very surprised at how fast this processed so I ran it again, 1st with differeent tolerance, then again with .0001 tolerance...same result as before 5 min. 39 sec.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • 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

×
×
  • Create New...