quote: NON oc'd at 2.44mhz it's around 1:24 if I recall.Out of the box an i7 920 runs @ 2.66GHz.

quote: continually staying in "caveman technology" will lessen its marketability What are you talking about??...I'd like to see you build a roman style door in Bobcad!! BTW..since your Govenator was on the advertising campaign...possibly you can get him to help "git with the program".

quote: You know you can get a sata to esata cord and just specify that its esata in the bios right? Hella cheaper than a card. Yes, but you still take up the slot in the back of the machine. Some Esata enclosures (add your own HD) come with the adapter...but since I run Asus the MB's comes with it...and USB 3.0 and Sata III

Version 9 also had import GD&T chook, and reverse post (tape to shape) that they got rid of in X, both BAD moves to get rid of, they are useful for their intended purpose. If you have an old NC program and no source file...reverse post was your only ticket to shop support without redraw.

The Corsair PS is very good, I just built a home machine with the 650 here. Not sure on that mobo, but if EM says so...so be it. I have a X58 1366 machine at work with the i7 920 and I'm running the Asus P6T Deluxe V2 board and its very stable and trouble free at this point (2 months). You can add the Sata III thru the PCI at a later point since that board doesn't come with it. Only other thing I'd want is Esata connection for fast external drive BU...but again a card can be added, didn't see that on that board.

The 930 is a LGA 1366 socket mobo so memory is tri channel and all 3 need to be the same. So when you want to upgrade the memory the 1024's will all need to be replaced. I'd go with 2 gig sticks (6g total) and be done with it. Plus side that motherboard should support future 6 core processors. Do you know the power supply and which motherboard it comes with? If you have a choice get a Sata III ready board then upgrade to Sata III SSD (as they are on their way out now)...and future proof. Price is fair and that system would run Mastercam very well.

quote: How big is the benefit of moving to the 1800?Well if you believe the Win 7 video benchmark the FX580 will get you 6.4 and for 2.5 x the price the FX1800 will get you to 6.7 ($160.00 vs $400.00) If your recommending a budget build the FX580 should be more then ample for 90% of Mastercam users (IMO). The 580 is faster then then old FX 1700 was...by a good margin. As far as building a $2,000 PC, you can build a very fast i7 with an 1156 socket 860, Asus P7P55D-E-Pro MB with Sata 6gb USB 3.0, Quadro FX580, 4 gigs G.Skill DDR3 1333 ram (to start), and good quality Corsair 650HX 650W PS, for under $1100.00. (I've built 2 now). Take the extra $$ and buy a new sata 6gb (Sata III) SSD (that are on their way) with 350 MB/s reads and 250 MB/s writes and you'll load MC in about 2 secs and read 50 meg files in near a blink.

I like to create surfaces on another level then analyze dynamic on the small "looking" radius surfaces (wireframe on at another color helps). I change those area surfaces to another color that stands out like red or orange (on large parts) so I dont forget to come back and detail them to size with the smaller cutter. If you have vericut you can just go see whats left with stock to highlight the smaller rads.

quote: For him to make use of the graphical improvements at a higher speed, does he need to upgrade to X3 or X4?So, you would suggest your client change to X4 to make use of the graphics card you advised...and spend thousands to upgrade and go thru a learning curve to make use of a $150.00 card?? Tough sale IMO. I've run V9 and X3 X4 under XP and a Quadro 570 and 580, and graphics were plenty fast. When I went to Win 7 V9 had bigtime issues with loosing curser, and you cannot disable hardware acceleration in win 7 with that card. Other then that, I say there is another bottleneck. possibly V9 doesn't have open GL enabled in config from previous video?

Building tooling in solids and using solid layout to get your 4 drawing views with hidden line removal or shown dashed saves alot of time. Also with solid layout you can extract very clean wire frame of part (similar to using SilhouetteBoundary chook except cleaner).

This can happen from time to time when you choose arc center and not an actual point. I've also seen it even more often on choosing an arc center as an entry point for pocket, when you regen it looses the point and defaults to X0Y0. This bug has been around a real long time and since its intermittent, Mastercam doesn't use the software enough to find it... its not been fixed for many years. I've seen enough parts scapped over the years to watch for it, and use actual points...the system will not drop those.

quote: I cannot use the roughing STL to verify the remaining finishing ops without getting an error,Have you tried to reduce the speed / quality toggle before running and then saving the rough STL? This toggle will make a huge difference in STL file size, and if its just rough path drop it all the way down to speed to reduce file size---then save STL and run finish.

I've not had that issue, but I usually will avoid steep transitions anyway. The only real tip I can give is use the smallest ball e/mill possible that wont break down...say a 3/16 w .007 x .007 edge break.

quote: Thanks guys. My application is more for deburring than anything and I'll give it a try. For that you might try locking to 3ax and use ball e/mill in 3D chamfer mode by just chaining the edge curve. Works great.

quote: Just wondering if there is and way to program 3d without having the software.Yes..it is possible if you have nothing but time on your hands and aren't really interested in making $$. "Back in the day" I've seen 3-D programs generated with an excel spreadsheet (before 3-D cad cam was widely available). Expect to spend the better part of a week to do what can be done in and hour by comparison.

quote: Just did some checking here on some surface paths i am doing on a mold, watched the second hand at the most 1.5 sec to open the operation. On the Core 2 DUO I had at the time it was 3+ seconds. Keeping in perspective I can open a small Mastercam file (double click file to open a second app that X4 says no to) takes less then 3 seconds on a Core i7 920 win7 x32.

quote: Seems to be related to the new style operation parameter page.Yep...I agree. I refuse to run X4 because of the added irritation of the delay, and the hodge poge of old style mixed with new slower paramater tree that forces a pull down to view all the params so they can leave it in a tiny window. Also the temp file error and lost toolpaths that would not regenerate cost me a bunch of time that helped loose faith in X4. These issues just added to the experiance (feeling) of a VERY SLOW system that is extremely irritating to work with in a fast paced evironment. Happily sticking with X3.

This is another issue that has never been corrected. You can set the correct stl in job setup (path and all) and its useless since upon restart it doesn't read and load (or retain) the file. It's been that way for years, (not just X4)...irritating lack of attention to "detail" IMO. Also the "auto jump" to the next operation group that resets graphics view (if group uses another wcs) after verify, and moves the operation manager marker to end of file is another rediculous glitch thats been there forever.

Depending on how much curve is in the floor often you can set the maximum gap size to an amount larger then the hole, then gap motion to direct and it will force the tool to stay down and cut right across the hole.

quote: I consider it the exact opposite. Unless you've got a $500,000 machine and a low budget programmer, the programmer usually costs more than the machine.$500,000. was propably lower then the average cost of a spindle at the last shop I was at, with 1 spindle at $2,000,000. Its not just the machine cost. The footprint cost for the machine tool, the material cost that sits and its footprint, the cutters and support tooling footprint all add to WAY more money then a computer, desk, programmer footprint with 20K worth of equipment. That will way more then offset the extra 5-7 $ per hour for a top programmer vs top setup person. Were not just talking about a programmer vs a machine. We are talking about cycle time that will usually mean machine + operator vs programmer. Does the programmer spend 2 hrs to save 1 hr cycle time? In many cases the answer should be yes. Point being, it's not always cut and dried to do things in the least amount of your time only as #1 consideration.

quote: Saving time and doing as little work as possible is my rule of thumb!Depends on material being cut and how many parts you are running. You have to consider spindle time at 2-3 times the value (cost) vs programmers time when making these decisions. If that were a "production part" and anything more then 5 minutes or so cycle time I'd hand drive all that toolpath by drawing GEO...Xform / offset and trim to cut near ZERO air cutting. Cycle time comes first...my time second for production IMO. Edit...woops just read the inside is deeper??? WTF??? Do the blue first then inside deeper 2nd.

quote: BTW, I noticed how you did not comment on the graphics in Esprit compared to MC. Espirt is still not as good as solidworks, but way closer than MC, way closer.That would be because last time I ran Esprit was about 17 years ago (early 90's). At that time a rectangular shaped deep pocket part with a kicked out floor took 2 days to toolpath in Esprit. I did a projected toolpath pocket to floor surface in around 2 hours and after the supervisor saw the difference that was the end of Esprit. As far as graphics goes, I have no issue with graphics on X3 with a Quadro 580 and win7, and to be honest eye candy is toward the bottom of my priority list...more a bonus feature IMO. I do understand why some are switching, and cant blame them when options are limited. Just dont expect across the board gains, because from what I hear, then only people switching feel forced into it from lack of options created by Mori. I consider this "bundled software" stuff kind of like buying an external hard drive with built in backup software....I re format and use the good stuff.

quote: I still think it's psycho to have that stuff residing on a Machine Control.It doesn't take a good shop owner long to do the math on a $200,000 +++ machine tool sitting while the guy programming goes to see if they have a 1/2 ball 3.0 OOH on hand as opposed to a $15,000 cad system sitting. Not only that...often after the operator learns a little something they develope the greener grass syndrome and want to be a "big bucks" programmer full time. It wont stick IMO...MC will put their 64B code on hold and get posts working to take back whats doing exit stage right as we speak. quote: It will always be necessary to create additional geometry for constraints and boundaries.I agree, just a matter of how much and how easy is it. As long as there is clamps, tool tabs, bolts, tombstones and other fixture / tooling and components to consider in toolpath, plus forgings, waterjet profiles, and misc. shapes to save material being implemented, there will be a need for addition GEO....period.

quote: when dragging arrow to a point on the surface and the arrow will snap to that exact point ...Looks like your right about that, it will snap to an actual point, pretty sure it didn't used to when I first went looking for line / normal some years back. Just goes to show...you have to try something different when you get fed up things are taking to long the old way.

quote: even easier, use Create Point- DynamicProblem with that is dynamic is basically "sketch mode" and doesn't allow an exact point to input like project will. I use that when close is good enough..ie evaluating part. If you ever get a model file with points (pilot locations) only on a 5 ax part the project works awesome for drilling. quote: I have always done it like Rickster Dont feel bad...that option used to be under create point, surface project (to create a line LOL)...and even less people found it there. This is the improved place to create a line normal.