USB Audio DAC mod + LaserBoy Correction Amp

[ElektroFreak]

I don't know that we're talking about the same thing, but in the settings page there is a drop-down box just above the blanking optimization area that contains the ILDA test pattern..

Also, does anyone know if wet noodles will work in the interim until carmangary is able to help me with spaghetti? I have a private show that I'd like to do, but I can't use anything but DMX with my scanner right now since my copy of Spaghetti is not working properly ATM. Also, I'm at work, so I can't test it out here. Any info on wet noodles would be greatly appreciated.

[mixedgas]

LFI player may work for you.

its hosted at sourceforge.

Steve

[ElektroFreak]

Downloaded it and will try it out tonite. Many thanks!

[buffo]

My scanners are SP20's, which are rated at 20 kpps for the 20 degree angle I'm scanning the ilda test pattern at.

This is not correct. You can *not* scann the ILDA test pattern at 20Kpps and 20 degrees with those scanners! You will get all sorts of artifacts, including tails on the circle like you mentioned earlier.

The ILDA test pattern is designed to be scanned at 8 degrees. When scanner companies release informataion on maximum scan angle, they are not talking about the ILDA test pattern, they are talking about the maximum travel of the scanners. The ILDa test pattern is an extremely demanding frame that has many points which are spaced way too far apart to be scanned correctly. This is intentional, as it forces the scanners to run at maximum acceleration. (for example, the center "circle" is actually a polygon with 12 sides that is position completely outside the center square. But since the scanners never reach those points, they trace a ballistic arc instead, which forms the circle you see.)

Those scanners should be able to handle the ILDA test pattern at 20Kpps and around 9 to 10 degrees scan angle. No more. Reduce you scan angle and take another picture.

If you re-tune the scanners to a slower speed (say, 12Kpps), then you will be able to scan wider. But of course, you'll get more flicker as well.

For example - I'm running the Scan Pro 50 scanners. I re-tuned them to 30K speeds, and now I can scan the ILDA test pattern at 18.5 degrees and still have the pattern look perfect. But if I re-tune the scanners for 50K speed, then I can only scan the ILDA test pattern at 6.5 degrees. Since I'd rather have the wide scan angle at 30K than a narrow one at 50K, I keep my scanners tuned to 30K and always run them at that speed.

Adam

[ElektroFreak]

^That explains why my ILDA test looks wonky, as I'm definitely scanning at much more that 8 degrees. Since I finished playing around with the blanking optimization settings, I've been able to get my system looking pretty much 100% perfect with the original EzAudDAC driver. My final settings were: Frame Start Hold: 20, Light Hold: 8, Blank Hold: 0. With these settings, the system works great for both animations and beam shows. I couldn't be happier.

One note, however: I have noticed that when I alter the scan angle using the red square in the "scanner settings" form, the stability of the image goes way down. Instead of a steady image, I end up with one that wobbles and twitches slightly. The same happens when I alter the "H skew" and "V skew" settings. The only way to get back to a stable image is to uninstall Spaghetti and delete the "Harper Ingenuities" folder from the Program Files directory and reinstall. To change the scan angle, I use the volume controls for the USB sound card. Changing the Pitch and Yaw settings has no effect on stability and are perfectly safe to use.

This may only occur on my system, but I thought I'd mention it in case any other people have seen this issue.

[drlava]

Glad you are reaching a point that is acceptable for you. Have you tried the ilda file that I posted above? Typically, you shouldn't use the volume control to adjust scan angle because that will also dim your lasers but for occasional testing it works OK.

[ElektroFreak]

I tried messing around with that new driver, but honestly in my case I had more luck with your original. The settings I tried didn't seem to make things look better, and in a couple of cases made them look worse, so I just went back to the original.

I'll continue trying different settings with the new driver using that .ild file so I'll see if I can't come up with a setting that works even better than what I've got now. Also, I haven't noticed any excess dimming when using the volume controls on the soundcard vs. the scan angle adjustment in Spaghetti.. Once again it could just be working well on my system, so I don't know if the same holds true generally or not.

[DPSS]

Here's how it looks with my projector with the sound card DAC system. I noticed that if I tuned the blanking offset to the ilda pattern, it made a tail on the 3rd circle but with the ilda test pattern blanking 'detuned' it looked like the attached photo.

Try out the attached ilda file with optimization off (you can put it in the Spaghetti test frames folder.) Start with small scaling and scale up till the arcs are just inside the circle markers. Have the driver blank delay set at 6 and post a photo of the result, if you are willing. This might help reveal blanking issues if you have any, and I'm interested to see what it looks like on your system.

Ok, I created the 50fps show with your file and followed your indications. This is the result:



I hope the quality is good enough, I'm affraid I couldn't take a better picture.

Still, after reading Adam's post, I think I'll try retuning my scanners. For the moment they are factory tuned, which I understand is suposed to be quite good for SP20's, but better safe than sorry.

This is not correct. You can *not* scann the ILDA test pattern at 20Kpps and 20 degrees with those scanners! You will get all sorts of artifacts, including tails on the circle like you mentioned earlier.

The ILDA test pattern is designed to be scanned at 8 degrees. When scanner companies release informataion on maximum scan angle, they are not talking about the ILDA test pattern, they are talking about the maximum travel of the scanners. The ILDa test pattern is an extremely demanding frame that has many points which are spaced way too far apart to be scanned correctly. This is intentional, as it forces the scanners to run at maximum acceleration. (for example, the center "circle" is actually a polygon with 12 sides that is position completely outside the center square. But since the scanners never reach those points, they trace a ballistic arc instead, which forms the circle you see.)

Those scanners should be able to handle the ILDA test pattern at 20Kpps and around 9 to 10 degrees scan angle. No more. Reduce you scan angle and take another picture.

If you re-tune the scanners to a slower speed (say, 12Kpps), then you will be able to scan wider. But of course, you'll get more flicker as well.

For example - I'm running the Scan Pro 50 scanners. I re-tuned them to 30K speeds, and now I can scan the ILDA test pattern at 18.5 degrees and still have the pattern look perfect. But if I re-tune the scanners for 50K speed, then I can only scan the ILDA test pattern at 6.5 degrees. Since I'd rather have the wide scan angle at 30K than a narrow one at 50K, I keep my scanners tuned to 30K and always run them at that speed.

Adam

Thanks for the explanaition Adam, it clarified things very much for me.
Before doing any more tests, I'm goint to try and retune my scanners as best as posible using HE Laserscan's 12k ilda test pattern and your tutorial this weekend and go from there.
Any last minute suggestions you feel I should take in to account?

Domingo.

[drlava]

Thanks for testing it, at what point rate did you do that, sir?
It tells me that your tuning is off, and this is causing blanking issues at the point speed that you played that file at. So you are on the right track to retune, 'stage 0'. After the re-tune, then we can go after blanking with that file.

[DPSS]

No problem, thank you for all your help . The speed was set at 20kpps.
As soon as I get done with the tuning I'll post back.

Domingo.