quote: Why is this that in 9.1sp2 when I am in true solid verify on mill,it seems that after I verify my part,if I decide to hit the "Screen Fit" button,my verified part totally dissappears and then when I go back to my geometry screen,the geometry looks all screwed up! I've noticed this a couple of times. What I'll do depends on what I was doing. If I just ran verify and I want to see the result, I'll hit page-down a few times until my geometry shows up. If I just went in to verify, I'll exit verify, do a screen-fit, and orent things the way I want using Dynamic rotate, then try verify again.

You can also go to Screen|Config and change the config file from the metric to inch. After doing that, MasterCAM will ask you if you wish to re-scale. It is a cool feature.

quote: I am using the original file that was installed wi SP2. Also, my dealer was notified and suggected deleting the config file and restarting MC to reset the config files to its default. No luck. He is not able to fix the problem.Deleting the .CFG file won't do anything to the .TXT file, and it's the .TXT file that contains the messages. I'd suggust taking a look at the MILL9.TXT file before you go futzing with drivers, mainly becasue the .TXT file is easy to check and fix. The 166-169 messages are located at line 10,040, and the 436-439 messages are located at line 5543 (as counted by CIMCO Edit).

Without being able to see the file, it's difficult to know what the problem might be. The non-manifold body error on boolian ops is usually related to the kinds of problems others in this thread have already mentioned. It is also possible that the solid is damaged, though that seems somewhat less likely. Well, unless you trimmed one of the solids with a surface. That seems to be a real bugbear for any solid modeling package. Before I tried drawing it again, I'd try using the NOHIST C-hook (was that part of 8.1?) to remove the history tree from the solids in question and re-do the boolian. If you don't have NOHIST, export the solids as Parasolid files and re-import them. They'll be dumb solids w/ no tree then, and you can try the above. If the model(s) in quesiton do invovle surface trims, look at those trims and see if there isn't some way you can defer the trimming operation(s) until after you boolian the models.

It sounds like you are trying to use the MILL9.TXT file from MasterCAM 9.0. The missing messages that show up when you go into Screen|Config are the strings associated with the new converters - Rhino, CATIA, and Unigraphics. When the config dialog comes up, it reads those messagees from the numbered list in MILL9.TXT. If your MILL9.TXT version dosn't match the version of MILL9.EXE you are using, you'll get errors like that. Probably a few others, as some of the other new fuctions will map to other 'missing' strings. Try one of the following to fix: - Restore the original MILL9.TXT file for V9.1sp2 - Copy the MILL9.TXT file from another V9.1sp2 intall over your busted one - Delete the MILL9.TXT file and do a 'Repair' install of MasterCAM from the MasterCAM CD.

What I'll do in that situation depends on what geometry I'm trying to add. If it's on the same plane I am working with, I'll set my Z (I'm kinda compulsive about setting my Z) and have at it. If it's on a different plane, I'll set the WCS back to Top, do my stuff, and set it back again.

SCSI is nice in a workstation becaue it off-loads disk processing to the SCSI controller. An IDE system will pause momentaraly when accessing the disk. SCSI-based systems don't, and that makes them feel much more responsive. The only downside I have noticed is the cost. quote: One thing I can tell you is NEVER MIX SCSI AND EIDE DRIVES (if your scsi controller is on the mobo) or you're in for a nightmare of a time keeping it running good.I have done this on a couple of occasions. Don't necessaraly reccomend it, but it dosn't cause insumountable problems, either.

Looking at the specs for the Dragon 2 board, I see that it has a seperate RAID controller and supports serial ATA RAID, so my comments about CPU performance don't seem to apply to this board. That Windows is noticeably snappier is a common consiquence of such arrangements. Any time you off-load disk I/O to a dedicated controller you gain a lot in system responciveness - that's why I still like SCSI for high-end workstations. A note of caution here though: RAID 0 is less reliable than any other RAID configuration, as the loss of any drive will trash all the data. Being a belt-and-suspenders kind of guy, I only use it for scratch/capture drives where the data won't stay long. On a home/game system that will get upgraded often and lives in a climate-controlled environment, this is not a *huge* concern. Looks like a killer MoBo though. Every possible bell and whistle. A bit spendy, perhaps, but the best often is.

Non-specific threats are near useless when applied in a small geographical area. When applied to a country as large as the US, they are worse than useless becasue they needlessly divert scarce resources in responce to some undefined threat. i.e. if the government knows that there is a plan to attack Chicago, it's kinda silly to have first responders in podunk little burgs 2,000 miles away spend time and effort preparing for it.

Motherboard-based IDE RAID implementations are generally steaming piles of poo - to use a technical term. No cacheing means you are using the disjointed caches on the hard drives themselvs. No dedicated I/O coprocessor or I/O bus means that the CPU has to do all the RAID management, and all the intra-RAID traffic has to move hither and yon over the PCI bus. The best you can hope for under such conditions is a small loss in performance. To do RAID right, either IDE or SCSI, you need an add-on RAID card that will implement the RAID level you have chosen in it's own dedicated hardware. FWIW, starting this year I'm specing serial ATA RAID for new servers/NLE/video capture systems using Adaptec's serial ATA RAID boards. Not as many channels as SCSI offers, but the servers I build rarely have more than 5 spindles anyway.

RAID 1 is mirroring. The second drive is a copy of the first drive. RAID 1 provides good data protection, but consumes half of your installed drive space. Read performance can go up a bit (depedning on the quality of the implementation), but write performance suffers. RAID 0 (striping without paraty) can provide a real boost in performance (depending on the implementation), but makes the whole drive subsystem less reliable. Personally, I'd not bother with RAID on a CAD/CAM workstation. It adds complicaiton without providing any tangible benifit. CAD/CAM applicaitions are computationally intensive, but are not big users of disk I/O. i.e. they are computaitonally bound, not I/O bound. Bigger returns on your dollar would be found in a faster CPU or more RAM.

quote: May be solids memory management not fully related to the database allocations and big toolpath allocations are not leaving place for solid memory management especially when U work from Solid faces.I think you smacked it on the head there. Solids in MasterCAM are running out of the parasolid kernal, and likley using dynamic RAM allocations as other windows programs are. Toolpaths and the like are using MasterCAM's fixed allocations. Bump the fixed allocaitons up enough and there won't be enough RAM for the dynamic stuff.

I encountered some issues early on when trying to use the WCS the same way I used the UCS in AutoCAD. I find that my results are more consistent if I use cplanes for drawing the way I always have and reserve the WCS for the toolpaths that require them. Working coordinate systems are real useful for toolpaths, but less so for geometry creation. With that said, you can still use the view manager to retain cplanes you will be using again. Go through whatever process you are using to create a WCS, just don't actually change the current WCS.

Create|Next Menu|Polygon. The dialog box that comes up has help functions and a picture that changes depending on the options you select.

If you have to remove and replace the key for the server to see the NetHASP, there is something wrong with that paralell port. Suggustions: - Don't plug anything into the back of the NetHASP. - Make sure the port mode is set to ECP or Standard in the computer's BIOS. EPP or EPP+ECP nearly never works. Don't set it to bi-directional, either, as the industry never did seem to aggree on what that meant. - If the above ain't getting it done, or you need to connect something to that port, consider adding one of these to the computers in question: AladdinCard They are high-quality add-on paralell ports that hide *inside* the computer, making them harder to get to. Not too spendy, either. About $30.00US, last time I checked.

Glad it helped. Rights in general can be difficult to troubleshoot. Rights in the registry are even more troublesome. Developers generally cannot determine what effect unfortunate combinaitons of rights might cause, so the error messages related to rights difficulties are almost always misleading. Generally what I'll do in a CAD/CAM lab is assign the users the same registry rights as granted to the Power Users group. This does not make them Power Users, as there are several other user rights that they need to be able to take advantatage of that access. Generally it's only HKEY_LOCAL_MACHINESoftware that needs adjusting. Be careful not to inadvertently change the SYSTEM user's rights to any part of the registry. That can cause very strange and difficult to track down problems.

The hasp state is stored in user section of the registry. Specificly, it's stored in HKEY_CURRENT_USERSoftwareCNC Software, Inc. as the REG_SZ value Sim Type. Local would be 'H' while network would be 'N'. If I had to guess, I'd say that the registry on the computer in question is corrupt, and it's corrupting the USER.DAT copied from the user's roaming profile location when the user attempts to make a change to the HASP settings. Another possibility is that the NHASP9.EXE program on that computer is damaged. Once the USER.DAT file becomes corrupt, things that depend on the settings in the unreadable section won't work. Like MasterCAM. Try using REGEDIT to take a look at the HKEY_CURRENT_USERSoftwareCNC Software, Inc. key and see what happens to it when you use NHASP9.EXE to change the HASP type.

Have you taken a look at the Solids book on this website? Lin and Shue have a Solids book out too, though I haven't yet had a chance to take a good look at it.

quote: Only one taker eh? Sheesh, I thought for sure I'd get at least 5.I'm starting at San Bernadino Valley College in the spring [1], in addition to the teaching I'm doing at Glendale College [2]. And some interesting things are happening at work, too. It looks like I'll be off this summer though (budget cuts). If you are still looking for someone then, let me know, 'k? [1] Teaching Hydralics. [2] Teaching MasterCAM.

quote: Say I want to cut a pocket, is there an easy way to offset the tool within the machine controller (as opposed to modifying the mastercam program) to cut the pocket bigger? Not without editing the code to add G41/G42 moves on your own. MasterCAM's pocket roughing does not use in-control comp, and there is no way to force it on. If you want to change the size by just a little bit, you can turn comp on in the finish pass and add enough finish passes to clean everything up.

What Iskander teh Lazy said + What tolerance do you need to hold? How deep is the hole? What kind of machines do you have avalible to you?

quote: Sometimes when I'm working on a project that requries me to move my geomerty around alot, I have found that my circuler geomotery will likely to turn into splines.I'd need to see the before and after files to tell for certan what is going on. I can say that most projections and some translations will result in splines instead of arcs depending on the exact location of the source and destinaiton.

quote: This may or may not help but it reminded me of a similar problem we used to have. Is there any chance the original ellipse was created in Autocad? We had the same facet problem trying to machine those but found that if the ellipse was exploded before export, then it cut fine.AutoCAD ellipses from AutoCAD R13 and later can only be exploded if PELLIPSE is set to 1 to force the creation of polyline ellipses. Ellipses have been native AutoCAD objects since R13.

This bike: replaced this bike: Not sure if either of them count as toys though. The '99 Sprint ST was this many when I got rid of it this past June: The Speed Triple has ~20,000 miles on it as of Friday.

quote: Why would I not be able view .SLDDRW files in Mastercam, when I can convert the .SLDPRT files? Wouldn't it be easier if we could convert both extensions in Mastercam. .SLDDRW files contain no 3D solid information. All they contain are links to the .SLDPRT file that the projected views were generated from. If MasterCAM converted them, all you'd get would be a somewhat difficult to work with print. Paper would be better.