Let's discuss this Chinese printer driver

[john banks]

If we can produce our vectors (including text converted to vectors) in .ai format and our bitmaps in .bmp format we can open RDCAM/Laserworks to automatically import these.

I have installed a generic Adobe postscript printer driver and in the port setting pointed it to a file. Then when I print from a windows application I get a .ps file produced. (alternative would be to make a CUPS driver which I've only just started looking at).

Using a Ghostscript script, ps2ai.ps, we can convert the postscript file produced into an ai.

The issue so far is that this ai contains the bitmaps and RDCAM likes to have the bitmaps separately, so they show as a black square.

Postscript as many are aware is a page description language, and just stripping out the bitmaps so far is eluding a simple solution.

Other possibility is to print to pdf to see if it is easier to get the images out of that with the command line.

The other option is to instead dump the printed file into CorelDraw which is quite easy, then it is one further click to get it into RDCAM, with the existing plug in separating out the vectors and bitmaps for us.

BTW, this is all intended to be automatic, so when you click "print" from a windows application, the result would appear with no further intervention into CorelDraw or RDCAM depending on how we do it.

Ideas, thoughts and advice appreciated.

[Ross Moshinsky]

I think you're going to have to combine the VBA script for CorelDraw and a Generic Print Driver from Windows to get the results you want. I know very little about the subject but I'm fairly sure that is the way forward. You might even need to look into combining a Generic Print Driver with a Plotter Print Driver so you can get the logic behind vector cutting added.

I still don't understand why the Chinese don't develop their own printer driver. The current way of doing things makes no sense. I'd pay a good bit extra to get a machine with a proper print driver. After dealing with proprietary software for the last 10 years, I'd like to stay away from it as much as possible.

[Vicki Rivrud]

I've been using Lasercut since 2004, upgraded to 5.3 at this point. It outputs direcectly to either Corel X4 or AutoCAD 2000. I can import .ai files as long as compression was not used in saving the files.

Forgive me, but from a users perspect I do not understand why using the Chinese software to drive your project to process versus the capability to just "HIT" print and have a windows "print driver" print it to the laser like a printer is such an issue. It is just an additional step to open the controller software and import the file, set the settings and send to the laser machine or usb stick if a stand alone.

I get this question so many times, as we specilize in Chinese lasers, specifically the Leetro Controllers. I train others and offer tech support to our customers, who otherwise would not receive support from China as well as being a moderator in a Chinese Laser Support forum.

I've attended a few "American" machine seminars and watched how they send files "to print" and you still have to setup power,speed, numer of cuts etc. So I am really puzzled why so many want a "print driver" instead of just using the controller software, which does just that.

I look forward to your opinions. I am not versed in RDCAM, however it is still the same concept.

Vicki Rivrud
JR Laser Solutions

[john banks]

There is actually an open source project on the Leetro (Lasercut) boards to do this, and a commercial solution.

The advantage of using a print driver is that you do not have to worry about software versions (how about X5 and Win 7 64 bit running Lasercut, does it work?) and choice when outputting your design. If you use design software that doesn't have a plug in then you have to export your design and then import it into your laser machine software. Often you have to split into vector and bitmap, and then import twice and realign them too.

There is widespread appeal, and rightly so IMHO, for a solution where you can output your design from any Windows software than can print. It will open RDCAM with your design, and if you'd set your layers correctly before you could hit go, but still with the option to alter more machine specific stuff not pertinent to your design like you do in the advanced tab of a print driver.

It becomes agnostic to software choice and versions.

Best of all would be an open source control platform which some think would raise the quality quickly.

Rewriting Lasercut or Laserworks would be a bigger job, although if you had the rld or mol file formats you could just export straight to the laser and never see the Chinese software at all. I don't think Laserworks is something I mind seeing though, but streamlining the user/designer to drop their design straight into it from any software is the idea. Is that appealing?

How about the workflow if I used Inkscape or an svg viewer on a webpage and I want to laser it? What if I hit "print" from the application and then it appeared in Laserworks and if I was happy I then hit "start"? Two clicks vs how many?

[Ross Moshinsky]

Some of us have files structured for different software. It's a HUGE pain when you have thousands and thousands of files structures one way and then have to change to another process for no real reason.

Example: With a mainstream laser, you can layer objects on top of each other, send it to the machine to raster, and it does the job. No issues. With a Chinese laser, all of the layers need to be combined and clean. The paths need to be essentially perfect. To add to this idea; customer supplied art for print is rarely perfect. It could take an hour or so just to get the artwork right. That's $100 in shop time. Not exactly something that can be passed on to the customer very easily.

Print drivers are superior in every way to the current method the Chinese machines use. If the Chinese used a print driver, I'd probably have 2 machines sitting in my shop right now. I'd happily pay $1000 extra per machine for a proper print driver (as long as I wasn't charged that amount directly) for a machine with a proper print driver. The versatility and the ease of use with a print driver vs using some Chingrish piece of software is immense.

John: I think you're wasting your time making a fancy file conversion. If you can't get rid of the Chinese software all together, I think it's quite pointless. That's my opinion at least.

[john banks]

The print driver is an (easier) first step that could be a stepping stone, but if you haven't used RDCAM with rewritten English labelling you might be underestimating how good it is and how well it handles layers, and in some ways it is less fussy about things like line width than others. It doesn't crash, it is very fast, it isn't bloated, is reasonably well laid out and has few more niggles than most of us would find with Corel or Adobe. It doesn't look wildly different to what you see in the advanced tab of some print drivers, with more options beneath the surface which you don't have to use.

Personally I find the biggest hassle is the exporting from ANY software and then importing. The extra clicks and multiple application launches, browsing files etc, remembering the version of the file you just made, sorting by date etc. You know the stuff. Or "click and it is there".

If other people don't find this it probably isn't worth taking it much further, but I think many do.

[Ross Moshinsky]

John, out of curiousity, are you starting from fresh or do you have a folder full of jobs you've already created? Believe me, it makes a difference.

Also, as I said, when a customer provides a layered file with everything stacked on top of each other and there could be 10 paths stacked, you'll find that working with RDCAM difficult. I know because I have the same problem with Gravostyle. The difference is, I can open it in Illustrator or Inkscape and send the job directly to the laser with no fuss.

One button to convert and drop it in RDCAM will probably be beneficial but for me, the real issue is still the fact you can't send it directly to the laser without going through RDCAM.

[john banks]

The job count on our laser is about 800 so far but we've not formally started commercially yet as we're still developing products, but we are getting customer submitted files and apart from the expected trouble with bitmap quality, compression, color, contrast, size etc, and having to hunt for or make vectors we aren't seeing any trouble so far. One company we've been developing ideas with (but no sales yet) tell us they are interested because our quality is much higher than their existing supplier who appears to put no effort into optimizing their files and just dumps things to their laser like it is a printer, and no amount of drivers will change the human thought needed.

Are you saying with that background that making a printer driver that dumps either the whole thing as a raster into RDCAM, or dumps the vectors and rasters into RDCAM would be unhelpful (I would like to give the choice)? What killer feature is it that we're missing is putting you off? Can you explain what it is about the setup you like that sorts out the stacked paths as we may be able to incorporate it?

Last night I send the windows test page to my postscript driver and then imported it into CorelDraw and then exported it to Laserwork. There were LOTS of layers. I could see that in the four color windows logo they'd actually made it up of lots of slightly different coloured lines which all produced their own layer in RDCAM. It looked OK, simulated OK and didn't crash, but it would seem sensible to organise it automatically first.

If you were sending the Windows test page to a laser on which you like the setup, what would you like to see? Dump it all as a raster image with dithered grayscales to represent the colours? If you printed this file to your laser as vector, what would you like to see?

If you tell me the actual features that you think would improve it I could try to distill that into a solution. At the moment I'm trying to work out what you do and don't want to see. For some files you'd see an RDCAM window and then simply hit start (or download if you don't want the laser to physically start).

Some people complain about the idea of printer drivers because they don't think they have the control from their design software that is more relevant to the laser.

I think what I want is flexibility with what software is used to output and removal of exporting/importing manually.

I should also point out that I'm not doing the design or running the business. I'm just technical support to try to produce a nice system for my wife to use and contribute to the community, but insights from her as she actually trades, but before that you guys who are experienced traders, are invaluable to try to improve the workflow. I can without trading myself see room for improvement in the workflow offered with a Chinese machine. I also don't want to spend a year rewriting the entire software that talks to the laser. If I present the software with something it can digest I think it will be very usable.

If you can post an example of 10 layers stacked I can see how it is handled, but so far I've had no trouble at all with crazy numbers of layers converted to raster by RDCAM. If it is an issue and you want raster than Ghostscript and others can do it no probs.

[Ernie Balch]

I print directly from Corel X4 with my Rabbit laser. Everything seems to work about the same as the Epilog driver demo I got except, there is no library of recommended settings. I am also limited to using XP, not a huge problem, since I converted to Macs when Vista came out.

It is easier than exporting a bitmap and then a vector file to Lasercut where I had to make sure they were aligned properly.

Anyone else following the progress of the lasersaur open source laser build project? This system looks like a copy of the Chinese lasers but uses different electronics and software.

Their approach is to make G-code based laser cutters. No mention of how they would do raster files, but G-code is compatible with my CNC router.

ernie

[john banks]

I went to see a Lasersaur based build this week. He uses a web interface on his Mac and opens an svg file and then hits go and it downloads to an Arduino. It looks interesting, but in many ways the workflow is worse. He alters the G-code in a text box to change speed and power. I think he has to navigate to the svg he wants to use in web interface rather than just hitting a button inside illustrator and having it appear. If it was a complex vector with multiple layers I think it would be much tougher than RDCAM. The lack of a control panel on the machine itself I would miss. Presently he didn't have it rastering and it might not raster well and fast on an Arduino. He is interested in a board and is looking at RDCAM, but doesn't like PC only, doesn't like closed source and rightly can suggest may ways to improve things like the zooming/panning in RDCAM. But then in an Epilog print driver do you zoom and pan over your work at all?

Can anyone run me through their ideal scenario of a printer driver? Would you rely on the user of the package to tidy up their layers and cut order before hitting print? Is the way Epilog do it seen as ideal? It looked a bit lacking in control to me I have to say.

[Ross Moshinsky]

It sounds like you don't have a lot of experience with the other lasers. I don't have much experience with the Chinese lasers but I do have a lot of experience with the way they process their files.

With a Chinese laser it's all about paths. Everything has to be a path. Not only does everything have to be a path, everything has to be a perfectly closed path. There is no stacking a white shape on top of a black shape on top of a white shape on top of a black shape and having it all work out perfectly. If you work with logos enough, you'll find out that they stack colors A LOT. This is because when you go to fill spaces with the paint bucket tool or just about anything else, you get little white space around the edge. You get a MUCH better look with just adding a shape slightly larger behind the object. For traditional printing it's WYSIWYG so it doesn't matter how perfect the layers and paths are.

For me, the laser driver should be run much like any printer driver. If you want to raster something, it should be WYSIWYG. It should not care about paths at all. For vector cutting, it should be equally simple. If you see this color and it's this stroke, vector cut/engrave it. Simple. Obviously within the printer driver GUI you should be able to specify your speed, power, turn auto focus on and off. Specify DPI. Specify the orientation of the page and whether it is flipped or not. Turn on or off air assist. The basics of running a laser. Then have a button where you can get into the more advanced settings.

Like I said, Gravograph handles files very similarly to the way RDCAM does. I've found several different work arounds to get the job done. One that I default to is simply circumventing Gravograph all together and running the files straight from Illustrator. The WYSIWYG way in my opinion is by far the best way to handle this type of work. Going the paths direction makes life so much harder than it needs to be.

[Rodne Gold]

WYSIWG is terrible for other machinery , I use my Corel files for CnC engravers , overhead routers , digital printers and vinyl cutters - all use other back end drivers/programs to run the machines
If you design using Corel on a wysiwyg basis, you have to do a huge amount of extra work after to use the files in other applications. I think rdcam as a driver is way more potent than my GCC printer drivers , albeit the GCC's are easier to use for a low level operator.

[Ross Moshinsky]

WYSIWG is terrible for other machinery , I use my Corel files for CnC engravers , overhead routers , digital printers and vinyl cutters - all use other back end drivers/programs to run the machines
If you design using Corel on a wysiwyg basis, you have to do a huge amount of extra work after to use the files in other applications. I think rdcam as a driver is way more potent than my GCC printer drivers , albeit the GCC's are easier to use for a low level operator.

Engravers, routers, and vinyl cutters are all completely different beasts than laser engravers. The whole point of a laser engraver is the fact it's basically a raster printer with the capability of following a vector path as well. If you're prepping a file knowing it's going to be used for a CNC machine or a vinyl plotter, then you spend the extra time to make it work with those programs. If you aren't, then the extra time spent preparing the files for being universal when they will never be used in those other applications is just wasted time.

I'm suggesting more capability not less. Right now the Chinese machines only allow you to work with perfect paths. I'm suggesting if you want to work with perfect paths, go ahead. If not, you can still have "sloppy" artwork and get the job done. I know about 10 different tricks to create perfect path artwork from sloppy artwork but it took years and various pieces of software being developed to get to that point. It's still extra, unnecessary steps considering a laser can and should be able to raster based on the WYSIWYG principle.

John has obviously already mentioned that the issues of importing/exporting is what's driving him to go this route. That's another very valid point which I completely agree with. Add in the fact the artwork needs to have essentially perfect paths and proprietary software and you'll see why I'm not a big fan of the Chinese method of sending jobs.

[john banks]

If you like your wysiwyg view of a multilayered and filled/shaded vector logo then you are going to raster it. The existing Corel plug in will keep it as vector, unfilling it and giving each shape an outline and import multiple layers. This is the wrong method to import to RDCAM, it needs to be imported as a bitmap which is presently something you have to do manually but could be automated +- dithering.

Is this the issue Ross?

[Ross Moshinsky]

Here is a very simple example of a file that will never go into RDCAM natively. Can it be converted to work in RDCAM? Absolutely. The question is, why do I want to waste the extra time doing it? Not to mention, this is something I drew up myself just now. As I've mentioned, sometimes you'll get a file where the paths are not closed perfectly. A place where you will see this is on details in a logo. For example, where I made the interior of the "D" black, someone might not close that path. I've seen where a logo had a tree with hundred+ leaves and each leaf was an incomplete path which used stroke and fill to create the look the artist wanted. Then figure they also layered everything on top of each-other (sometimes 4-5 different colors) and it becomes a huge mess. If I remember on Monday, I'll send you a perfect example of a file like this.

I know it's possible to work around what I'm talking about. I've done it for a while now. It's just something that's annoyed me for years. It's a complete waste of time and energy.

226032

[john banks]

Thanks for the file. You are intending to raster this presumably.

RDCAM is not setup to receive filled vectors. If you're going to raster it why not send it as a bitmap? That is effectively what say an Epilog print driver does when you select "raster". Ghostscript can do the same with the vector output in a postscript file.

It seems straightforward and here it is in RDCAM. There is one extra step involved compared to an Epilog which is "convert to bitmap" in CorelDraw. A second button could be added to the CorelDraw plug in which forces everything to raster.

You could easily have two print drivers for the Chinese machines or an option in the print dialog like Epilog have. Send it to the raster only print driver if you want it rastered. Otherwise use the one that keeps the bitmaps and rasters that were already there.

226033

Does this solve the issue?

[john banks]

I'm new to VBA for Corel (but not to VB.net) but it isn't hard.

Just record a macro and then look at the code:

Sub Macro1()
' Recorded 03/03/2012
Dim OrigSelection As ShapeRange
Set OrigSelection = ActiveSelectionRange
Dim s1 As Shape
Set s1 = OrigSelection.ConvertToBitmapEx(cdrBlackAndWhiteIm age, True, True, 500, cdrNoAntiAliasing, True, False, 95)
End Sub

[Ross Moshinsky]

Thanks for the file. You are intending to raster this presumably.

RDCAM is not setup to receive filled vectors. If you're going to raster it why not send it as a bitmap? That is effectively what say an Epilog print driver does when you select "raster". Ghostscript can do the same with the vector output in a postscript file.

It seems straightforward and here it is in RDCAM. There is one extra step involved compared to an Epilog which is "convert to bitmap" in CorelDraw. A second button could be added to the CorelDraw plug in which forces everything to raster.

You could easily have two print drivers for the Chinese machines or an option in the print dialog like Epilog have. Send it to the raster only print driver if you want it rastered. Otherwise use the one that keeps the bitmaps and rasters that were already there.

226033

Does this solve the issue?

RDCAM works with 72-90dpi screen resolution. So any time I convert a vector to a bitmap, it's going to turn into a relatively low resolution image right off the bat. I essentially have to export it at 500dpi (which will make the image roughly 6x the size) and then scale it down once again to keep the high standard of the vector image. This is what I call a work around. Natively in the print driver, I assume it automatically converts it to a raster image, but it keeps in mind the raster resolution selected and the file size reflects that.

I think what you're saying will probably work very well and get the job done, but it's not at the level of the US laser's print driver. To me, if you're going to do this, it should work very similarly to a Print to PDF driver. They allow a lot more adjustment to keep the quality of the export high if desired or very low if desired. Maybe that would be the happy medium.

[john banks]

1 is an ai file loaded into Corel. It is a reasonably complex vector. You can see the size in mm. I click one button which runs a macro which drops it into 2 at the same size, reduced to 1 bit, dithered and at 500 DPI as you can see from the zoom in 3.

This dither pattern at 500 DPI is too tight to get good contrast but that is a matter of selecting the correct dither to apply in my macro, this is only a first test/proof of concept.

What do you think? Your thoughts have been really useful to sort out this issue and I think we're getting there? I can't see a loss of resolution.

If your layers are set correctly, to raster this vector logo is two clicks of the left mouse button away.

226039226040226041

[Ross Moshinsky]

Any non-black color you have flexibility. It's impossible to ask for really clean edges on those colors because of the way the laser handles the image. Under those circumstances you're absolutely right that the high resolution of the image will have an impact on the images dithering.

Where I'm concerned is if the image is black and white. Under those circumstances, are you going to get the very clean edges? Are you going to get the crisp details? That's where I see some issues in your example and have seen some general issues with exporting a vector as a raster image. The only way I've found around this is to export at 300-600dpi and then scale it back down afterwards. The edge quality and level of detail does get effected by working with screen resolution images. It's not a HUGE issue but it's an unnecessary loss in quality.

[john banks]

226043226044

Here it is at 1000 DPI. You can see the detail, scale in mm. It took 48s between Corel and Laserwork to convert this vector and hand it over, but 144 Mpixels is not exactly small. More sane sizes take a few seconds, no more than they do with a print driver to prepare for printing on a conventional printer.

[john banks]

I'll try a curve at 1000 DPI next, in black. I'm expecting pixel perfection...

[Ross Moshinsky]

I don't think 1000dpi is necessary. 300-600dpi should be more than acceptable. The big issue I see from these examples is the fact that black is not solid black. It probably won't effect anything but you'd like to see if you start with a true black vector image the black section is solid black.

I think the real test is text. Working with a 1"x3" logo with text is where you will see the limitations of this process vs vector. When it's working off a vector file natively, you know the limitation is just your machine. When you're doing it this way, you're adding the extra variable of how well the processor does.

In the end, I will take back what I said about me not seeing this as very useful. This is clearly a reasonable way of doing things. With a bit more tweaking it will probably prove itself very useful.

I forgot to mention this yesterday but another reason I like a true print driver with no RDCAM is for printer sharing. With a "real" printer driver, you can share the laser on a network easily. This way doesn't change that. Sharing the laser on a network can be pretty useful.

[john banks]

226045226046

100mm circle at 1000 DPI. Zoomed in you can see a perfect edge with 0.025mm pixels.

There is no loss of quality I can detect. I can throw other files through the conversion, but I have no reason to suspect they'll fail. It is just that we can control the mechanics of how the vector to raster conversion occurs rather than it being hidden inside a proprietary driver. This may be a positive depending on how you look at it.

The examples were probably CMYK and not true black and that needs some playing with in the color mapping and dithering.

I'll do some black text in Corel at 1000 DPI in RGB black and post...

Agree on the network. But that is not at all impossible although I won't be doing it.

[john banks]

Arial 72 point, showing "72" and a close up of the curves. 1000 DPI. Looks perfect to me.

226047226048

[Ross Moshinsky]

http://www.brandsoftheworld.com/logo/youtube-1

That would probably be a good logo to test. If I were prepping that to engrave, I would fill the red with solid black and engrave it at 500-600dpi. With your method, I'm curious to see how the text would hold up at 300-500dpi if the logo was around 3" wide. If it can keep the text quality up, your method might be very effective.

[john banks]

The logo is quite shaded, it is a question of style whether you want to force it to undithered monochrome or try to dither. If you want to quickly put together a file that is still vector then I can throw that at the macro.

The dithering needs more work and like you say you end up with bits that are supposed to be black that end up dithered and I don't like that. But I think the text if black and white vector will come out very crisp in bitmap up to the limits that it can do for the resolution it is rendered at.

[john banks]

Here is the logo at its original size simply thrown at the macro, so dithered.

226051

[john banks]

Here are some close ups. mm scale.

226052226053

[john banks]

This one has the "You" fill changed to black and the area around "Tu" fill changed to black.

226054

[Vicki Rivrud]

In Lasercut, I can send directly to the laser. I have so many other legacy pieces of equipment, cnc router, vinyl cutter, large format 36" printer, laser printers etc that Windws XP Pro is on all our workstations.

HMMM . . .very interesting comments and I do understand that it would be great to eliminate the import step. I guess in my training and experience - being "sloppy" wasn't an option, no disrespect intended to anyone.

I just make sure my file is prepped correctly/fully etc before hitting Lasercut, then if something may have been missed I can always use the unite line option in lasercut.

It seems to be quick and easy for me if I prep well to begin with, yes it takes time but IMHO it has proved to be worth it for our projects.

Vicki

[Ross Moshinsky]

If someone is ordering a $50 item and 25% of that is material costs. That leaves $37.50 left in labor or 22.5 minutes to do everything else. Taking 5-10 minutes to rework artwork doesn't exactly fit wage structure more often than not. For me to fix the mistakes of others to gain no profit and not improve the end product is simply wasting money.

John,

If you send me your print driver in a version that simply produces a PS file (so no import) I'll happily do some quality tests for you. I won't be doing it on a Chinese laser or in RDCAM but my software works similarly to RDCAM and lasers are mostly just lasers.

[john banks]

Thanks Ross, but it isn't at the stage yet that testing would be useful as the actual printer driver is written by Adobe, it is a generic postscript to file, and as you would expect it works perfectly, but I haven't packaged it sensibly to give you an installation as you need to mess around with custom .inf files to make it work on Windows as it comes as a .ppd. The glue needed in this project is to pick up the postscript file and convert the vectors in the .ps file to .ai or .bmp, and the bitmaps in the .ps file to .bmp with dithering. Yesterday's discussions had me adding code to the existing CorelDraw plug in to handle rastering a vector as we discussed which is actually more immediately useful.

Going back to the questions in post #1, they are still outstanding about the best scheme to tackle this. We quite like Corel and improved functionality of the RDCAM plug in will come first given progress yesterday (adding the code I showed).

[jason harris]

This looks like in interesting development so I will keep an eye out.

I don't have my laser yet, it will be on a boat this week so I will have a couple things to learn. I am not new to CNC machines and the workflow from what I have seen seems similar to what you are doing on the chinese lasers vs the 'printer driver'.

I think it can come down to what you are used to. Ross, taking your $50 example I don't think I would be wasting too much time and loosing money fixing up vectors. People have been doing it for a long time outside of lasers and most of the time it is done with a click of a button in software. After all, doesn't this printer driver have to include it's own logic to do the same thing? It can't be out of the question that other software has this capability outside a printer driver.

For what I need a laser for and the return I would make out of it I don't think a laser that comes with a printer driver would be economical. It looks like those european ones cost a heck of a lot more than the chinese ones.

When I do get that $50 job the money for labour I make goes in to my pocket. The chinese laser allows that. If I got a european one with a printer driver it would be a long time before that money started to make it's way in to my pocket rather than paying off the machine.

As I say, I don't have my laser yet so zero experience with a workflow including lasercut but it doesn't seem very different than a workflow on a cnc router/mill as an example.

Also, if you are going to factor in job's and costs etc then it should be noted that most in my experience are very easy and only some are harder. Not every $50 job is going to give me a headache, only a few and if that is the case then it's not a $50 job. I dont have to do it.

As I say, it's what you are used to and I don't think it's as big a burden as you suggest to people that are doing it all the time.

[Ross Moshinsky]

We do $50-100 jobs all day. 22.5 minutes to take an order, design the file, run the job, assemble the job, box, and take payment is nothing. We've been at this a while. You don't add on time if you don't have to. The fact is, if you outsource the work, it costs $5-15 per logo to clean it up. That eats up just about all the profit on a $50 job.

And artwork clean up is not a click of the button. You will learn soon enough. There is a lot of work involved in getting clean vector artwork. That's why it's so common for graphic designers to just layers things on until they get it right. Think of a painter. They don't just slap paint on a canvas and it's perfect the first try. They have to constantly go back and rework. Many graphic designers are no different. Just the other day we had a client come in and had a tree which had gradients and layers and layers. It was not simple to rework into a file that had perfect paths for importing. It was something that could be run using the WYSIWYG method right from Adobe Illustrator though.

[jason harris]

I think you may have missed part of my point.

No, as I said I have not worked with lasers so I may be misunderstanding something fundamental until I get going with mine. I am not new to cad, working with vector based software or cnc machines and cnc software which also complains when vectors are open on a lot of operations.

There are a lot of things that can be done with one click to fix up things. Often vectors that are not closed are nearly closed and not noticeable until zoomed in on. This kind of stuff can be fixed up fairly easily with software.

Having said that, you will use your laser for different things than I will as mine will be another tool used for more general engineering and I am sure that you will have hassles with drawing that I don't ever bother with.

I think the point still stands that the approach may be different but I don't see it as a big time waster and a lot of people will be using lasers for the kind of thing I will do while some people doing what you do find your procedures better.

Unless you are competent in both workflows I don't think you can understand the pros/cons of each and say one is a universally better approach.