Results 1 to 10 of 10

Thread: New laser amateur with some questions

  1. #1
    Join Date
    Sep 2021
    Posts
    5

    Default New laser amateur with some questions

    Hi everyone!
    Im pretty new to show lasers, and I am fortunate to find this awesome place of information!
    I've had a Cameo Luke 700 for about a year now, running in auto-mode on my events. As I wanted to try to do some more programming, and wanted a brighter laser, I bought a Laserworld DS-3000RGB, which has integrated shownet, and a reasonable price for my humble needs.

    I'm trying to do some comparision to see which one is "best". I understand Laserworld is not "state-of-the-art" but maybe its enough for me. I dont care much for animation, I like dreamy beams overhead... (The beams is more crisp on the Cameo, but the galvos actually shake the picture alittle more.)

    Anyway, couple of questions arise, as I like to understand how things work. Sorry if somebody has already asked it before.

    1) The DS-3000 comes with a set of ilda-files on a SD-card, which can be played automaticly. I notice that some of the animations/pictures has some "blanking errors". Like there is an extra 5cm line between two figures that is not supposed to be there. Now, if I reduce the default speed(20k) DMX-slider down to an area about 10k-15k it shows perfectly, but if I instead increase to 30kpps, the unwanted line gets even longer. Other (simpler or more complex) ilda files may or may not have this error.
    So is a specific ilda file "ment" to be played at an exact/predetermined scan-rate to be showed correctly?

    2) I also note that especially red color is alittle more prone to the error mentioned above, than the other colors. Maybe thats the tradeoff when not going high-end-lasers...?

    3) I do see laserworld web-pages and manual are warning alot about not driving the galvos too fast at too large scan angles. The default on my scanner is 20k, and its rated for 40k. Hopefully I should be able to show most animations at full 50 degree angle without being afraid of killing my poor galvos? And intermittingly increase to 30k when needed.


    Thanks!

  2. #2
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Quote Originally Posted by Sqvirrel View Post
    I'm pretty new to show lasers, and I am fortunate to find this awesome place of information!
    Welcome to PhotonLexicon! Always nice to see new folks. When you get a moment, be sure to update your user profile with your location. That helps people replying to questions, because often times the answers are different depending on what country you live in. It also makes it easier for you to link up with others who may be local to you.

    Also, I apologize in advance for the wall of text that follows, but you asked some good questions. I wanted to be sure you had good answers.

    I'm trying to do some comparison to see which one is "best".
    That depends on your criteria for "best". As you pointed out, the Cameo has a crisp, tight beam. Since it's only producing 700 mw, it almost certainly uses single-mode diodes, which will give you a very tiny dot. This is great for graphics and abstracts.

    The Laserworld projector is making 3 watts; it's using multi-mode diodes. That means the spot on the wall is going to be a good bit larger. Thus, when you are displaying graphics, the image might look more "fuzzy" with that projector. But for beam shows, the Laserworld projector is going to appear much brighter than the Cameo.

    Which one is "best" depends on what is more important to you.

    Referring to the Cameo, you said that "the galvos actually shake the picture a little more". I'm not clear on what you're referring to here. Do you mean to say that the image flickers more? Or does the entire image jitter and shake, as if the projector itself were being jostled while it's running?

    Flicker in the image is caused by trying to scan a detailed image at too low of a scan speed. The fix is to reduce the complexity of the image (fewer points), or increase the scan speed. But, of course, your scanners need to be able to handle the higher speed, or you can damage them. (Overdriving scanners will cause them to overheat, which will damage them, but there's also a risk of them slipping into a resonance band, which can be catastrophic. If you ever hear a screeching sound coming from your scanners, REDUCE THE SCAN ANGLE IMMEDIATELY.)

    If the entire image is shaking or jittering around, then that is a different problem. It could be caused by noise on the inputs to the scanner amps (possibly a loose wire, or a bad ground connection), or it could be caused by trying to scan the image too wide, causing the scanners to distort the image as they are pushed beyond their capabilities. If you are seeing the entire image distort like this, then immediately try reducing the scan angle to something very reasonable, like 8-10 degrees. If the image stabilizes, then you know you were scanning the image too wide. (Pro tip: Don't scan too wide! Over time you will overheat and damage your scanners. And if you accidentally drive the scanners into resonance, you can quickly destroy them.)

    I notice that some of the animations/pictures has some "blanking errors". Like there is an extra 5cm line between two figures that is not supposed to be there.
    This can usually be corrected by adjusting the "blanking offset" setting. However, when you are playing content that is stored on the SD card, you may not be able to change this setting. But when you have the projector connected to your PC using the Shownet software and controller, you should be able to adjust the blanking offset. Usually a setting of 3 or 4 points is optimal. (The setting refers to the delay, measured in points, that is applied to the color signals, compared to the X/Y position signals. The color signals are delayed to allow the scanners to "catch up", since the lasers can be turned on and off in microseconds, while the scanners can take several milliseconds to change direction.)

    If I reduce the default speed(20k) DMX-slider down to an area about 10k-15k it shows perfectly, but if I instead increase to 30kpps, the unwanted line gets even longer.
    Yup. Makes perfect sense. Slower scan speeds allow more time for the scanners to get to where they are going, so they can catch up to the color signals. Thus, the mirrors are in the perfect position when the laser turns on and you don't have any tails. But as you increase the scan speed, you can run into a problem where the scanners aren't where they're supposed to be yet, but the lasers still turn on, so now you have this unwanted line.

    You can try to adjust the blanking offset to fix this, but if the scanners can't keep up with the higher speed, you'll never be able to get rid of the tails completely. Note that the scan angle also plays a part here. If you are scanning very wide, the scanners need to move quite fast to get to where they need to be. By reducing the scan angle, you reduce the distance the scanners need to travel, so they have a chance to get to where they need to be before the lasers turn on.

    Other (simpler or more complex) ilda files may or may not have this error.
    There is another factor that can cause tails to form: poor ILDA artwork. If the ILDA frame was not designed to include extra points at the end of a blanking run, around corners, or around locations where the beam makes an abrupt directional change, then you can get all sorts of weird artifacts. Most ILDA frame creation software is smart enough to add these points to the frame automatically, but if the frame was created with a very simple editor, or if some of the anchor points were removed by accident, then you'll see artifacts in the displayed image.

    All of this is related to the fact that the scanners are trying to move physical mirrors, and those mirrors have mass and inertia that must be overcome. When we say that a set of scanners can do "30,000 points per second", all that really means is that the scanners can display the ILDA test pattern properly at a scan angle of 8 degrees. It does NOT mean that the mirrors can move from full left to full right 30,000 times per second. (Not even close!) In fact, the bandwidth of a set of 30K scanners is closer to 2500Hz, and that's only for a small step (~ 3 degrees), and even then you will have about 3 dB of distortion between the position signal and the actual position. When it comes to moving the mirrors from lock-to-lock (call it 60 degrees), you're looking at a bandwidth of ~ 250 Hz.

    So is a specific ilda file "meant" to be played at an exact/predetermined scan-rate to be showed correctly?
    In a nutshell, yes. Nearly all ILDA artwork is designed to be displayed at 30,000 points per second. Stick with that setting and you should be fine. Also note that scan speed refers to both the software display setting AND the tuning parameters of the scanners. To run properly at 30K, the software needs to be set at 30K, the scanners need to be capable of running at 30K, and the scanner amps need to be TUNED to 30K.

    With everything set for 30K, if you slow the scan speed down in your software, in most cases the image will still look more or less the same, but you will see the image start to flicker more and more as the scan speed goes down. This is because it's being drawn fewer times per second. The flicker will quickly get bad enough that you'll get a headache just from watching the image.

    With everything set for 30K, if you increase the scan speed in your software, then as your scanners reach their limits the image will start to distort. If you continue increasing the scan speed, your scanners will begin to overheat.

    Note, however, that if you have high-end scanners that can keep up with the faster speed (and they have been properly tuned to that higher speed), then the image should look the same at the higher speed, with one final caveat:

    There is one other factor that can ruin your images even if you have very fast scanners and everything is tuned properly: it's the practice of "pulling points" in an ILDA frame. Early ILDA artists realized that scanners were very limiting, so they tried to optimize their ILDA frames to get as much detail as possible while using fewer points. One way to do this is to create rounded curves by intentionally creating an impossible path for the scanners to follow, rather than a more traditional (and closer) spacing of points that just traces the desired curve. They did this by "pulling the points", or setting the points so far apart that the scanners would never get to the first point they were headed toward before the next point was sent. Thus, the scanners would average out the path between these widely-spaced pulled points, tracing a smooth sweeping curve with just a few points in the ILDA frame.

    Obviously you can't do this for every feature in the frame. You need to be able to make straight lines and corners too. Also, when the scanners are tracing through a set of pulled points, they are ballistic. That is, they are under constant, maximum acceleration. This is dangerous, as it's easy to pull the points just a little too far, which can lead to overheating. But as long as it's only used for one or two small elements in the frame, it's not an issue.

    The problem with pulled points is that technology has finally caught up. Today we've developed faster scanners that could actually reach those far-flung points. So when you display some of those older frames at higher speeds with these new, amazing scanners, that smooth, sweeping curve becomes a jagged, pointy triangle instead. (Oops!)

    I also note that especially red color is a little more prone to the error mentioned above, than the other colors.
    This is due to the different response speed of the diode drivers. Some laser software offers the ability to adjust the timing of the color signals individually, but if you don't have the option in software, there is a hardware solution that works just as well: a color correction board. Basically the idea is that you delay the color signals to the drivers with FAST response times so that they match up with the slowest driver you have. Then all 3 lasers will turn on and off at the exact same time. DZ came up with the original design, but he sold everything to Stanwax, so you'll have to go through him to get one.

    I do see laserworld web-pages and manual are warning a lot about not driving the galvos too fast at too large scan angles.
    Yup. Good advice, too! Don't push your scanners to hard, or you'll kill them.

    The default on my scanner is 20k, and its rated for 40k.
    Are you saying that the default setting on your Shownet software is 20K? That's odd, because most software defaults to the industry standard 30K. That being said, I would suggest running everything at 30K. Even though your projector has scanners that are rated for 40K, they are probably tuned for 30K.

    An easy way to check is to bring up the ILDA test pattern at 30K and 8 degrees. (At 8 degrees, the pattern will be 14 cm wide and 14 cm tall for every meter between the aperture of the projector and the wall.) Is the circle in the middle just barely touching the sides of the center square, and it is otherwise perfectly round? Cool. You are tuned for 30K. (If you were tuned for 40K, the circle would be very large and would be completely outside of the center square, while if you were tuned for 20K the circle would be very small and nowhere near touching the sides of the center square.)

    Hopefully I should be able to show most animations at full 50 degree angle without being afraid of killing my poor galvos?
    No. Absolutely not!

    At 30K you are only guaranteed 8 degrees. However, in practice you can usually go a bit wider than that, because while the ILDA test pattern is a very demanding frame, most of your ILDA artwork won't be nearly as complex. But the physical limits of your scanners still apply!

    Remember that 30K does not EVER mean "at full angle". It means you can display the ILDA test pattern (which is a very demanding frame) at 8 degrees. And even then, there is significant distortion in the ILDA test pattern image.

    Remember that circle we were talking about above? Look at the ILDA test pattern in your ILDA frame editor and try to find the circle... You won't see it, because the points that make up the circle are actually spaced well outside the central square, in a 12-sided dodecagon! Those points have been pulled to create that circle, and the difference between where the points are (outside the square) and where the circle ends up (just touching the sides of the square) represents that 3dB of distortion I mentioned above.

    When you are running graphics content, a good rule of thumb is to bring up one of your most complex ILDA frames and then slowly increase your scan angle until you just start to see some distortion in the image. Back off the scan angle just a bit, and that is your maximum setting for graphics. Period. As in, do not exceed this setting when running graphics or abstracts. (Beam shows can sometimes be pushed even wider, because the frames are not nearly as complex.) And always keep in mind that if you ever see the image start to distort badly while running a show, back off on your scan angle. If you don't, you might find yourself buying new scanners.

    If you absolutely MUST have as wide of a scan angle as possible, you may consider reducing your scan speed in the software to 20K. This will allow you to scan wider, but unless you also re-tune your scanner amps to 20K you won't get the full benefit. Still, it might be good enough without retuning. Note, however, that by slowing down your scan speed you will get lots more flicker. And even at 20K you won't be able to go the full 50 degrees.

    Adam

  3. #3
    Join Date
    Sep 2021
    Posts
    5

    Default

    Thanks for a very good reply! I like to bury myself in the subject at hand.

    As I'm reading, some thoughts come to mind...

    When I play back a random ILDA file from the SD card (by selecting filenumber with a DMX-slider), I dont actually know what scan angle the galvos are operating at, and different ilda files have different width of the beams/animations too! Thats why I was thinking it was "safer" to leave the SD-card-playback-scan-rate-option to 20kpps. I do have the DMX-zoom slider which makes things smaller, but I still dont know. But maybe as you're saying, I should beware of resonance sounds, abnormal sounds from the galvos, or distortion, and I should (hopefully) be fine...

    Well, I have no Idea why it was set to 20kpps when I first checked the settings. But I can of course up it to 30, if that is more correct! Of course, when I play back different ILDA-SD-files I have to pair the pattern-number with a correct change of scan-speed, since many of the factory-ilda-sd-files apparantly plays with less blanking lines at lower scan speed...

    I still havent tried ShowNETs ShowEditor, I will do that, and check out the ilda test frame!

    Since I am mainly interested in slow moving overhead beams/"carpets", most of my projections show without any flickering at 20kpps. Is there any point in just display them at 20kpps instead of 30kpps to let the galvos relax alittel bit, and maybe retain some more longevity? I mean, is there any use of using 30kpps for a projection, if it anyways runs flicker-free at lower scan-rates?
    Last edited by Sqvirrel; 09-03-2021 at 23:12. Reason: Additions

  4. #4
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Quote Originally Posted by Sqvirrel View Post
    When I play back a random ILDA file from the SD card (by selecting filenumber with a DMX-slider), I dont actually know what scan angle the galvos are operating at
    There should be a global "size" setting that you can access. That will control the scan angle. You'll probably need to hunt through the manual for the DMX display board to figure out which channel is linked to this setting.

    I was thinking it was "safer" to leave the SD-card-playback-scan-rate-option to 20kpps.
    In general, you can scan wider at the slower speed, yes. The trade-off is more flicker in the image.

    I do have the DMX-zoom slider which makes things smaller
    That could be the master size control. But check the manual for a listing of all the DMX channels to be sure.

    I should beware of resonance sounds, abnormal sounds from the galvos, or distortion in the displayed image, and I should (hopefully) be fine...
    Yes.

    I have no Idea why it was set to 20kpps when I first checked the settings.
    You will often see the stand-alone DMX playback boards installed in less-expensive projectors with slow scanners. So it makes sense that the board defaults to 20K, because many lower-end projectors only come with 20K scanners. However, the literature for your projector states that you have 40K scanners. (Although in all likelihood they have been tuned to 30K.) Increasing the speed to 30K should be just fine - provided you don't scan too wide.

    Since I am mainly interested in slow moving overhead beams/"carpets", most of my projections show without any flickering at 20kpps. Is there any point in just display them at 20kpps instead of 30kpps to let the galvos relax a little bit, and maybe retain some more longevity?
    Running at 20K will allow you to scan wider, which will probably be to your benefit for beam shows. So long as you don't have any annoying flicker, slowing to 20K is probably a good plan, at least for your beam shows.

    Adam

  5. #5
    Join Date
    Sep 2021
    Posts
    5

    Default

    Thanks for the answers!
    There is still one thing I cannot quite grasp.
    I am projecting a smiley with the showeditor application. But the two eyes are not complete circles. More oval shape, and the lines are not closing.
    If I decrease the kpps from 30, it starts getting better. At 12kpps the eyes are finally around and everything looks like it should on the preview on the computer screen.
    What is this indicative of?
    Is it poor galvos that cannot keep up with 30kpps? Just tweak some settings to fix it? The ilda file is made that way?
    Since I am pretty new to lasers, and don't understand everything, its hard for me to tell what is poor quality or if something is underperforming/suboptimal, and what is normal behaviour (due to physics etc.).

    Btw, I have been looking at the ILDA test images. The 12kpps test pattern is looking quite normal, the circle is maybe alittle off, but I think that can be tweaked with settings (but maybe the scanners are tuned for 30 anyway?).
    On the 30kpps file however, the center square with the circle is so damn small on the wall, that it is almost impossible to differentiate if its correct or not! Am I doing something wrong here?

    I've come to realize that things are maybe not as simple as I first thought...
    In my simple mind, I thought that an ILDA-pattern was made with a specific/absolute size(angle) and kpps in mind. And therefore, when played back, if I didnt touch the size/zoom slider, the size/angle projected would be what the ILDA-file-creator intended?
    To take it further:
    All the stock patterns that comes with a laser, are already tuned to 30kpps and plays with the best and exact scan-width as intented by the pattern creator...
    Its not so? When I download an ILDA-pattern, I have to change to scan-width/size(or kppps) to somthing my galvos can keep up with?

    Regards
    Sqvirrel
    Last edited by Sqvirrel; 09-07-2021 at 08:38. Reason: Additions

  6. #6
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default Time for a little test...

    Quote Originally Posted by Sqvirrel View Post
    I am projecting a smiley with the showeditor application. But the two eyes are not complete circles. More oval shape, and the lines are not closing.
    If I decrease the kpps from 30, it starts getting better. At 12kpps the eyes are finally around and everything looks like it should on the preview on the computer screen.
    OK, we need to do some troubleshooting here. It's starting to sound like the scanners in your projector are only 12K scanners, or at the very least, that they have been tuned down to 12K. This would seriously suck, since the website advertises that your unit should have 40K scanners. To be sure, we need to perform a simple test.

    The 12kpps test pattern is looking quite normal, the circle is maybe a little off, but I think that can be tweaked with settings
    The 12K test pattern (technically, it's the 12/30K pattern, but I know what you mean) is the one most people use for tuning. It should look like this when it's being displayed properly:



    Technically there is a white box around the outside of the test pattern. In the image above, it's been clipped out (sorry), but this white box would form a border around the pattern right at the edges of the image.

    The 30K test pattern looks very similar, except that there is another, MUCH LARGER white square, that surrounds the entire pattern very, very far from the center border. But inside, you should see the exact same features that are present on the 12/30K pattern. It's possible that your projector can't display the large 30K pattern wide enough for you to view those center features very well, but they are there.

    The idea behind the large 30K pattern was that you could leave the scan size set very wide, and the relevant part of the pattern (the inside bit) would still be at 8 degrees or less. But there's no guarantee that this will be the case; some projectors might still scan too wide, while others might not be able to scan wide enough. This is why it's always better to use the standard 12/30K pattern when tuning (or testing) your scanners.

    Now we're coming to the test I was talking about. Please display the standard 12/30K ILDA test pattern pictured above using your Showeditor software. Start with a scan speed setting of 12K. Then adjust your scan angle so that the image is 14cm tall and 14cm wide for every 1 meter between the projector and the display surface. (That equals an 8 degree scan angle.) Just measure to the white border that runs around the outside of the pattern. Note that you don't have to have the projector at exactly 1 meter; it can be any distance from the wall that is convenient. All that matters is that the displayed image is square, and that it is 14 cm in each direction for every meter between the projector and the wall. (So if the projector is 2.5 meters away, the image should measure 35 cm on all sides.)

    Once you have this image displayed at the correct scan angle, start adjusting your scan speed in the software. If the circle is too large (bigger than the central square), increase the scan speed to make the circle smaller. If the circle is too small (not touching the sides of the central square), reduce the scan speed to make the circle larger.

    The question we need to answer is this: What scan speed setting in software yields an image on the wall that is closest to what you see above? (That is, at what speed is the circle just barely touching the sides of the center square?) Whatever that speed setting is, THAT IS THE SPEED YOUR SCANNERS HAVE BEEN TUNED FOR.

    Please perform this test and reply here in the thread with your results. Your answer will tell us a lot about your projector and what might be going on with the image distortion you're reporting. If you can take a picture of the displayed image while you have the scan speed adjusted for the best possible image, it would be very helpful if you could post that image along with the speed setting. If you can't take a picture, at least take notes about how the circle looks. (Is it perfectly round, or does it look more like a slanted oval, for example.)

    Adam

  7. #7
    Join Date
    Sep 2021
    Posts
    5

    Default

    So, I have 3 pictures here: 12kpps, 20kpps and 30kpps. All 3 pictures taken with approximately correct angle, 3,1 meter distance. At the wall, the projection was 2cm larger than in my calculations. If you think it is crucial I get it perfect, I can redo the projections with more accuracy(which means I have to dig up some better measurement tools in the shed!)

    So it seems the scanners are tuned for 20 then? Well, maybe it makes sense, since I was complaining in the first post that many patterns actually looked better in 20kpps?
    Anyway, as you stated, the scanners are advertised as 40kpps 8, so it is clearly room for improvement here... Have they done it like that to prevent noobs destroying their galvos the first day they get their new laser?
    Can they be retuned for 30?
    By the way, I havent done any of the laser optimizations guides on the laserworld site. Could do that to improve the shape of the circle, I guess...

    Thanks for your help so far!
    Attached Thumbnails Attached Thumbnails 30kpps.jpg  

    20kpps.jpg  

    12kpps.jpg  

    12kpps.jpg  


  8. #8
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Quote Originally Posted by Sqvirrel View Post
    So, I have 3 pictures here: 12kpps, 20kpps and 30kpps.
    There are 4 pictures in your post, not 3. However, the last two pictures appear to be the same image. Also, they appear to be in reverse order. That is, the first one looks like 30K, the next one 20K, and the last two appear to be at 12K. Is that correct?

    All 3 pictures taken with approximately correct angle, 3,1 meter distance. At the wall, the projection was 2cm larger than in my calculations.
    That's close enough.

    So it seems the scanners are tuned for 20 then? Well, maybe it makes sense, since I was complaining in the first post that many patterns actually looked better in 20kpps?
    From the pictures you posted, yes, it appears that your scanners are tuned for 20K. (That would be the second image, where the circle is touching the central square.) So that solves one mystery: The reason your projector has trouble displaying images at 30K speeds is because your scanners aren't tuned for 30K, they are tuned for 20K instead.

    Now, the question is: did they tune them for 20K because that's all they are capable of? If so, then you got screwed. The Laserworld website states that your projector comes with 40K scanners, not 20K. But your test image clearly shows that your scanners are running at 20K. Not cool. At the very least, they should have been tuned to 30K, which is the standard. Tuning them to 20K is just wrong, which is why I suspect that they are, in fact, only capable of 20K speeds.

    To figure this out, you'll need to open the lid of the projector and take a picture of the scanner amp PC board. Ideally we'd want to be able to read the markings on the board to get the exact model, but It's possible that we may be able to identify the scanners based on the general layout of the board. See if you can get some good pictures of the inside of your projector and we'll go from there.

    The other issue is that your blanking offset is not set correctly. Notice the two off-set horizontal lines near the top of the test pattern? The vertical "tick mark" should be in the middle of the offset space between the left and right halves of each line. Try increasing the blanking offset setting by one and see if this doesn't improve things. (You should notice the inside ends of the horizontal lines shift when you adjust the setting, but I'd also like to remove some of the tails that are showing at the top of the center circle, and the blanking offset adjustment may help that as well.)

    Have they done it like that to prevent noobs destroying their galvos the first day they get their new laser?
    Tuning a set of 40K scanners to 30K will already offer plenty of protection. Most people expect scanners to be tuned to at least 30K, if not to their rated maximum speed. De-tuning 40K scanners all the way down to 20K makes no sense, which is why I suspect they really are 20K scanners to start with.

    Can they be retuned for 30K?
    If they really are 40K scanners, then yes, they can be re-tuned to a higher speed. But if they are only 20K scanners, then no, you won't be able to tune them to 30K. (Or, more correctly, when you try to tune them at 30K the image will look bad, and you will not be able to get the circle to touch the sides of the central square no matter how hard you try.)

    I haven't done any of the laser optimizations guides on the laserworld site.
    I looked at the site, but I can't find the optimization guides you are referring to. Got a link?

    Adam

  9. #9
    Join Date
    Sep 2021
    Posts
    5

    Default

    Well, this gives me something to think about.
    If the scanners are, in fact 20kpps, then its false advertising, and I have a case with the seller, which in turn has a case with laserworld. If I open the scanner, I will problably void the warranty, so I will try to clear things up with the seller first, or ask laserworld directly.

    Anyway, I have tried to project some small size ilda-file at the wall, and it had no problem driving the scanners up around 40kpps, with rock steady picture without any artifacts or drawing-errors. I dont know if that showd anything about what its capable of.

    https://www.showeditor.com/en/tutori...rate-kpps.html
    https://www.showeditor.com/en/tutori...imization.html

    Im not sure about the worth of those two links.

    Regards
    Sqvirrel

  10. #10
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Quote Originally Posted by Sqvirrel View Post
    If the scanners are, in fact 20kpps, then its false advertising, and I have a case with the seller
    I agree. However, it could be that the scanners are just tuned to 20K. But the only way to know for sure is to open the projector.

    If I open the scanner, I will probably void the warranty
    I disagree. You can't tune the scanners unless you open the projector. You also need to be able to open the projector for routine maintenance (e.g. cleaning the optics). Granted, the average user may not feel comfortable doing this, but if the manufacturer says that opening the case will void the warranty, then they should also have a procedure in place for you to send the projector back to the manufacturer for them to perform these routine maintenance tasks for you.

    I will try to clear things up with the seller first, or ask laserworld directly.
    Be sure to reply back here and tell us how they respond!

    I have tried to project some small size ilda-file at the wall, and it had no problem driving the scanners up around 40kpps, with rock steady picture without any artifacts or drawing-errors.
    This may indicate that the unit does, in fact, contain 40K scanners and they are simply tuned down to 20K. You would need to perform a lot more testing to be certain, however. (But opening up the projector should tell you right away.)

    The process they describe is one that will safely allow you to push your scanners faster and faster until you reach a point where they start to distort. But I wouldn't call this the "correct scan speed". It's more like trying to find the maximum safe speed for a given scan angle, which is not the same thing at all.

    Also, they tell you to set the size at 50%, which probably corresponds to something between 20 and 30 degrees. Because you are scanning much wider than 8 degrees, you already know that you won't be able to display the ILDA test pattern correctly, so I'm at a loss as to why they suggest that image as one of the ones to use during the test. While most graphics frames are not anywhere near as demanding as the ILDA test pattern and will probably look fine even at 20 + degrees, the ILDA test pattern is absolutely going to look poor (at least the circle in the center will).

    In my opinion, they're going about this whole process backwards. The standard is to tune the scanners to 30K, set your software to 30K, and then start increasing your scan angle (by adjusting the size setting) until you start to see distortion in the image while displaying normal graphics frames. Once you see distortion, you back off on the size setting until the distortion goes away, and that is your max size setting for graphics. But you leave the speed at 30K, which matches the tuning of your scanners, so they will be performing at their peak efficiency at that speed.

    This link has some very good information about how to adjust the blanking shift, as well as the point optimization. I would first write down the current value for each one of these settings, and then go through the process step by step.

    Since we already know your scanners are running at 20K, I would set the scan speed in software to 20K before you start. Then work your way through the process. Once you get the blanking shift set correctly, you can change the point optimization values from zero back to their original settings to see if that helps things even more or not. But getting the blanking offset dialed in first will be a real plus. (All the better that the software appears to offer separate settings for red, green, and blue!)

    Once you finish this output quality optimization procedure, bring up the 12/30K ILDA test pattern at 20K and 8 degrees. See if the tails on the circle in the center of the ILDA test pattern are gone. Also, look at the offset horizontal lines at the top of the pattern; are the vertical tic marks centered between the interior ends of each line?

    After you've gotten things to look as good as possible at 20K, you can try cranking the scan speed up to 30K just to see what happens. Of course, the circle in the center of the ILDA test pattern will shrink (and may start to resemble a spiral), but you may also see changes in the offset lines at the top. This is why you don't normally go messing around with the scan speed. The idea is to tune for a specific speed and then always run at that speed. Otherwise there are too many other things that will need to be tweaked to get the image to look correct.

    If this were my projector, I'd yank the lid off and inspect those scanners. If they were 20K units I would demand a refund and to hell with anything they try to say about voiding a warranty.

    On the other hand, if they really are 40K units, then I would re-tune them for 30K, set the software at 30K, run through the "output quality optimization" procedure, and then leave all those settings alone.

    The last step would be to figure out how wide I can scan before I start to see distortion in the graphics. Once I had that number, I wouldn't exceed it. (Need a bigger image? Move the projector further from the wall.)

    Adam

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •