The dot mode correction in Lightburn is an excellent tool and in early versions of Lightburn Jason realized the problem of a 1 pixel signal producing a sausage shaped 2 pixel burn. This is well explained in the excellent video cited by HeritageNI. In those early days, Jason set the signal output to crop the beam diameter from each end of a pixel string. This worked fine for longer pixel strings but the problen this caused was to eliminate any single pixels in a dither pattern. At the time there were no diode lasers to consider, where the PWM instant switching causes this sausage effect. With a glass tube CO2 laser it is possible to balance the response speed of the HV power supply to speed and power such that you can get round dots for square pixels. Thus in making the single dots disappear we could not get faithful 1pixel = 1 dot replication. As a short term fix, Pass Through was implemented so that no dot processing happened on externally dithered images. Dot shape correction has now been reintroduced and it is a 60% fix for the sausage problem on RF CO2 and fiber lasers where the. beam produces a ROUND dot. However, diode laser do not produce a round dot. As perfectly illustrated by Darren in this video, the dots are typically oval(ish) with a 2:1 aspect ratio. Think logically about that situation and you will see that there is not HALF a pixel burn at the end of each pixel signal but a WHOLE pixel burn. Thus for a diode laser a 1 pixel signal generates a 3 pixel wide burn (long sausages!!!) No amount of dot correction is going to fix this inherent problem. Lightburn is an amazing program that tries to make things simple for new users and has the goal to be a one size fits all program for any laser type. I would say Jason has been about 98% successful so far but there will be perculiaritie for each laser type that cannot be catered for. This video demonstrates one such problem. Darren has set a resolution of 254ppi but he cannot produce pixels shorter than 0.2mm wide That in turn dictates that his horizontal resolution needs to be 25.4mm/0/2mm = 127ppi BUT at the same time his Y DOT height is 0.1mm ie a resolution of 25.4/0.1= 254ppi.. Here is the Lightburn dilema. IF you set the resolution to127 you get a PRESET line interval of 0.2mm but that is not what you want You want a line interval 0.1mm. Effectively you want to scan each line of pixels TWICE to create a 0.2mmx 0.2mm burn. That is THE best resolution Darren can ever achieve for true photo replication. This resolution will be OK for large images 300mm sqaure but will look "blocky" ar 100mm square. Darren has succeded in his goal of of showing that there is more to engraving that fiddling with Lightburn parameters and sliders. The diode burn shape is a physical fact that cannot be changed so what Darren is trying to achieve with his tests and experiments in this video is an acceptable visual compromise. The interesing thing about slate is that it is a BINARY material. Simply put, it is grey compressed sand. When you heat sand it turns to pale grey glass. If you burn the same place twice it does not change colour of the glass. Thus the material can only ever be dark grey and pale grey, just 2 colours (a binary material). The dither test pattern. shows this tolerance to overlapping X axis dots. . The grayscale dithered pattern has been set at 254ppi so only the distance between dots in the X axis is changing between 0 and 255. The changing shades of grey you see in the test pattern is an illusion created by your eye as it mixes the light grey glass dots with the dark grey background. At about 150, the 0.2 wide sausage burns are just about touching. Remeber that the 245 ppi line spacing (Y axis) is already correct for the dot height (0.1mm). So between 150 and 217 the suasage ends are gradually overlapping to exclude all the grey background, Thus beyond a certain point there is no perceived colour change. ("the white lmit"). This is one of those pragmatic compromises that this test appears to indicate and it makes sense to set the max white image level to match this white point. However, there are now futher complications.. The test pattern is a reflection of how different DITHERED shades of grey in an image will appear. There is NO uniforn gradation from black to the "white limit" so an unadjusted image will have all the mid tones "overwhite" and almost uniform. This means the image must be "fiddled with" to remove single and double pixels (say) to restore the correct black/white ratio to the burnt image. Losing pixels means losing detail and this is the juggling act that Darren is trying to perform here. I would say he has done a pretty good job considering all the negative facts he is forced to overcome.
Thank you Russ - you articulate so well what I am not able to detail in my video - I appreciate you taking the time to expand on those inherent problems we face using diode lasers for either reproduction with photos 🙏
@@dafyddgould9485 Hi This ia Russ from Sarbar Multimedia. I have to assume that you have a diode laser. The claims for your m,achines's spot size maybe optically true but these machines are not projecting images. Instaed they are trying to concentrate light INTENSITY with a conventional lens system and that creates issues that are beyond the scope of normal lens theory. First it's very likely that the smallest spot you ca burn into your material will be 3 to 5 times bigger than that defined in the machine spec. That burnt spot will NOT be a constant diameter. Oh, by the way, a diode laser usually burns an oval spot that can be 50% wider than it's height and that creates another issue when trying to replicate pixels. The burn diameter depends on A) How accurate your focus is. B) The %power (exposure time in the case of a diode laser) C) The speed, and D) The material. I will be happy to send you a test pattern so that you can explore what your macine is capable of . Remember that it is essential to know your machine's capabilities because the size dot you can burn defines the size pixel in your image. Iif your smallest dot is 0.2mm for instance then the best resolution will be 25.4 / 0.2mm = 127pixels per inch image resolution. . Her is a private contact form if you need more information. forms.zohopublic.eu/ndeavorlimited/form/K40XtreeemLaserCutterContactRussSadler/formperma/k2Cn0QN5ChpazfTMAUw25lZ-FKpjZa96TQWHjv3ntOg
Thanks for watching Ashra - yes, now that I better understand what is happening from the dither algorithm perspective I know I can get better control over my final product :)
Good question Ashra - if you think about it the etching of the glass is producing the lighter markings just like on slate - so I would say yes to inverting - however I haven’t engraved glass as yet - on my todo list 🙂
Thanks Tim - Photoshop is definitely packed with functionality and can do so much - working on some automation processes at the moment and saves tons of time
Hi Carol - not really sure what you mean - slate being the dark background as opposed to white. The image needs to be inverted to get a "positive image".
@@mrcraftsmith I’ve been trialling a lot of 3d images rather than photos, for 3d I’ve found that the image pops off the slate better, or the 3d effect is more pronounced, when not inverted.
Hi Michelle - the concept of a lighter dog or person would require the same processing conceptually but at the other end of the spectrum with the white point
Have you tried it in ImagR? They have multiple algorithms for slate. Also, I have been sanding my slate smooth first, then I spray paint them flat black which I think gives better detail and grey scale effect (greyer for white than the goldish of slate)
Thank you for watching - I have tried ImagR - I showed that example in my first video on slate engraving - and yes the results were great - this was more a demonstration to show how you can create that dither without the external software other than Photoshop and Lightburn 🙂
@@mrcraftsmith Ok Another tip I invented sorta. for slate and tile I fast dry them in an air fryer after spray painting at 300 for 5 minutes then cool for like 15 and ready to go no waiting 24 hrs to use.
What happens if you use Jarvis? I have typically used Jarvis instead of dither Also does speed make a diffrence to the dot length? I see you varied the power, how about speed? Love your presentation 3 & 5 were my choice too
Hey Dave - I could be wrong but my understanding is that Jarvis is just another dither algorithm - from my testing the dot size doesn't vary with speed, however depending on the complexity of design it can be less accurate at much higher speeds
I use Jarvis too, not a fan of grayscale as too much white despite adjustments. Jarvis with high power non inverted but high definition gives me best results.
thank for a great video, ive watched both of the slate videos and made my own grayscale test. Now trying to figure out to make a dither test file so if you have tips or files at hand id love the help. Keep up the good word greetings from Germany
thanks for the feedback - if you have made the grayscale test in PS - I'm pretty sure I cover how to make it into a dither file - what software are you using or have access to?
@@mrcraftsmith is there a way to tell wich one develops spots before engraving? Maybe there is something to lookout for like the converging lines or something.
His previous video on Slate engraving (The one with the Koala engraving on slate has an incredibly helpful section on this) I'd only ask @mrcraftsmith if you would mind sharing all the testing files you've created? (The white to black range, DPI etc?)
Great video and lots of info to take away from it. Is the file you have to get the white output value available to download at all? Trying to get to a point where I'm happy with my image engraving and think this may help.
Hi Craig - thanks for watching - here is a link to the file - tinyurl.com/mrcraftsmithslatedither - if you haven't subscribed to the channel and you feel I'm offering some value to you - consider subscribing :)
@@mrcraftsmith thank you for the reply and I have subscribed. The images that you prepared fully in Photoshop, when you opened them in Lightburn did you activate Pass-through or process the image again?
Thanks for very useful info. Do you use something to finish the coaster surface? Shine or matt spray or wax or…? Trying to find something that would increase the contrast..
I started out with imag-r but once I dug a bit deeper into Photoshop I have been using that - although I'm sure with more testing I would probably find the sweet spot with imag-r - I guess it comes down to whatever tools you have available to you and what feels gets you the best results - there's no right answer as to what is the best tool :)
The dot mode correction in Lightburn is an excellent tool and in early versions of Lightburn Jason realized the problem of a 1 pixel signal producing a sausage shaped 2 pixel burn. This is well explained in the excellent video cited by HeritageNI. In those early days, Jason set the signal output to crop the beam diameter from each end of a pixel string. This worked fine for longer pixel strings but the problen this caused was to eliminate any single pixels in a dither pattern. At the time there were no diode lasers to consider, where the PWM instant switching causes this sausage effect. With a glass tube CO2 laser it is possible to balance the response speed of the HV power supply to speed and power such that you can get round dots for square pixels. Thus in making the single dots disappear we could not get faithful 1pixel = 1 dot replication. As a short term fix, Pass Through was implemented so that no dot processing happened on externally dithered images. Dot shape correction has now been reintroduced and it is a 60% fix for the sausage problem on RF CO2 and fiber lasers where the. beam produces a ROUND dot. However, diode laser do not produce a round dot. As perfectly illustrated by Darren in this video, the dots are typically oval(ish) with a 2:1 aspect ratio. Think logically about that situation and you will see that there is not HALF a pixel burn at the end of each pixel signal but a WHOLE pixel burn. Thus for a diode laser a 1 pixel signal generates a 3 pixel wide burn (long sausages!!!) No amount of dot correction is going to fix this inherent problem.
Lightburn is an amazing program that tries to make things simple for new users and has the goal to be a one size fits all program for any laser type. I would say Jason has been about 98% successful so far but there will be perculiaritie for each laser type that cannot be catered for. This video demonstrates one such problem. Darren has set a resolution of 254ppi but he cannot produce pixels shorter than 0.2mm wide That in turn dictates that his horizontal resolution needs to be 25.4mm/0/2mm = 127ppi BUT at the same time his Y DOT height is 0.1mm ie a resolution of 25.4/0.1= 254ppi.. Here is the Lightburn dilema. IF you set the resolution to127 you get a PRESET line interval of 0.2mm but that is not what you want You want a line interval 0.1mm. Effectively you want to scan each line of pixels TWICE to create a 0.2mmx 0.2mm burn. That is THE best resolution Darren can ever achieve for true photo replication. This resolution will be OK for large images 300mm sqaure but will look "blocky" ar 100mm square.
Darren has succeded in his goal of of showing that there is more to engraving that fiddling with Lightburn parameters and sliders. The diode burn shape is a physical fact that cannot be changed so what Darren is trying to achieve with his tests and experiments in this video is an acceptable visual compromise. The interesing thing about slate is that it is a BINARY material. Simply put, it is grey compressed sand. When you heat sand it turns to pale grey glass. If you burn the same place twice it does not change colour of the glass. Thus the material can only ever be dark grey and pale grey, just 2 colours (a binary material). The dither test pattern. shows this tolerance to overlapping X axis dots. . The grayscale dithered pattern has been set at 254ppi so only the distance between dots in the X axis is changing between 0 and 255. The changing shades of grey you see in the test pattern is an illusion created by your eye as it mixes the light grey glass dots with the dark grey background. At about 150, the 0.2 wide sausage burns are just about touching. Remeber that the 245 ppi line spacing (Y axis) is already correct for the dot height (0.1mm). So between 150 and 217 the suasage ends are gradually overlapping to exclude all the grey background, Thus beyond a certain point there is no perceived colour change. ("the white lmit"). This is one of those pragmatic compromises that this test appears to indicate and it makes sense to set the max white image level to match this white point. However, there are now futher complications.. The test pattern is a reflection of how different DITHERED shades of grey in an image will appear. There is NO uniforn gradation from black to the "white limit" so an unadjusted image will have all the mid tones "overwhite" and almost uniform. This means the image must be "fiddled with" to remove single and double pixels (say) to restore the correct black/white ratio to the burnt image. Losing pixels means losing detail and this is the juggling act that Darren is trying to perform here. I would say he has done a pretty good job considering all the negative facts he is forced to overcome.
Thank you Russ - you articulate so well what I am not able to detail in my video - I appreciate you taking the time to expand on those inherent problems we face using diode lasers for either reproduction with photos 🙏
how do i find out if or how i need to do this?
@@dafyddgould9485
Hi
This ia Russ from Sarbar Multimedia. I have to assume that you have a diode laser. The claims for your m,achines's spot size maybe optically true but these machines are not projecting images. Instaed they are trying to concentrate light INTENSITY with a conventional lens system and that creates issues that are beyond the scope of normal lens theory. First it's very likely that the smallest spot you ca burn into your material will be 3 to 5 times bigger than that defined in the machine spec. That burnt spot will NOT be a constant diameter. Oh, by the way, a diode laser usually burns an oval spot that can be 50% wider than it's height and that creates another issue when trying to replicate pixels. The burn diameter depends on A) How accurate your focus is. B) The %power (exposure time in the case of a diode laser) C) The speed, and D) The material.
I will be happy to send you a test pattern so that you can explore what your macine is capable of . Remember that it is essential to know your machine's capabilities because the size dot you can burn defines the size pixel in your image. Iif your smallest dot is 0.2mm for instance then the best resolution will be 25.4 / 0.2mm = 127pixels per inch image resolution. .
Her is a private contact form if you need more information.
forms.zohopublic.eu/ndeavorlimited/form/K40XtreeemLaserCutterContactRussSadler/formperma/k2Cn0QN5ChpazfTMAUw25lZ-FKpjZa96TQWHjv3ntOg
#3 was the one that caught my eye as well. Lightburn did a pretty good job also. Great video as always.
Thanks for watching Ashra - yes, now that I better understand what is happening from the dither algorithm perspective I know I can get better control over my final product :)
@@mrcraftsmith Would you invert the image if you are engraving on glass as well?
Good question Ashra - if you think about it the etching of the glass is producing the lighter markings just like on slate - so I would say yes to inverting - however I haven’t engraved glass as yet - on my todo list 🙂
No3 Way to go my man. Thanks for the info great to hear the Aussie voice.
Cheers Peter - should be more Aussies on TH-cam 😀
Nice job Darren, I agree #3 is great, great video, thanks for the ground work and info
Thanks for watching Jim - happy to share as I learn 🙂
its number 4 for me i think ! great video mate thanks
appreciate you watching - thanks
Phew now I dont need to get ImagR as I have Photoshop and didn't want another software to work about. Great video
Thanks Tim - Photoshop is definitely packed with functionality and can do so much - working on some automation processes at the moment and saves tons of time
4 or 5 looks the best to me!
Thanks for the feedback Terry - it is definitely a close call 🙂
Great video thank you
You're welcome Clare - thanks for watching :)
Great video by the way, thank you so much!
Appreciate you tuning in 🙂
Great tutorial mate, well done.
You're welcome Andrew - thanks for tuning in
I actually like #4
Thanks for the input Ross :)
I must say, from what can judje from the video, that I may like number 4 better as it seems the dynamic range is wider, with the white being whiter.
Hi fitguy - yep a close call and hard to pick the definitive winner
No. 4
:)
I don’t like inverted results on slate? I find I get better depth and definition using un inverted but higher power.
Hi Carol - not really sure what you mean - slate being the dark background as opposed to white. The image needs to be inverted to get a "positive image".
@@mrcraftsmith I’ve been trialling a lot of 3d images rather than photos, for 3d I’ve found that the image pops off the slate better, or the 3d effect is more pronounced, when not inverted.
I lean to #4 & #5
thanks for the feedback :)
Hi would you edit the image differently in photoshop if this was a white dog or people please?
Hi Michelle - the concept of a lighter dog or person would require the same processing conceptually but at the other end of the spectrum with the white point
I agree #3
Thanks Merrill 🙂
Have you tried it in ImagR? They have multiple algorithms for slate. Also, I have been sanding my slate smooth first, then I spray paint them flat black which I think gives better detail and grey scale effect (greyer for white than the goldish of slate)
Thank you for watching - I have tried ImagR - I showed that example in my first video on slate engraving - and yes the results were great - this was more a demonstration to show how you can create that dither without the external software other than Photoshop and Lightburn 🙂
@@mrcraftsmith Ok Another tip I invented sorta. for slate and tile I fast dry them in an air fryer after spray painting at 300 for 5 minutes then cool for like 15 and ready to go no waiting 24 hrs to use.
What happens if you use Jarvis? I have typically used Jarvis instead of dither
Also does speed make a diffrence to the dot length? I see you varied the power, how about speed?
Love your presentation
3 & 5 were my choice too
Hey Dave - I could be wrong but my understanding is that Jarvis is just another dither algorithm - from my testing the dot size doesn't vary with speed, however depending on the complexity of design it can be less accurate at much higher speeds
I use Jarvis too, not a fan of grayscale as too much white despite adjustments. Jarvis with high power non inverted but high definition gives me best results.
thank for a great video, ive watched both of the slate videos and made my own grayscale test.
Now trying to figure out to make a dither test file so if you have tips or files at hand id love the help.
Keep up the good word
greetings from Germany
thanks for the feedback - if you have made the grayscale test in PS - I'm pretty sure I cover how to make it into a dither file - what software are you using or have access to?
maybe try this file - is designed for slate drive.google.com/file/d/1qal-eg1MIKSei1JHdN-U2CqRBTV2Oqem/view?usp=drive_link
for me the nr 5
Thanks for watching
Any Idea on how to get rid of the spots in the slate seen at 10:44 ?
Unfortunately not - that is the nature of the stone - hence why I used it for this test as I could use that and supply to a client
@@mrcraftsmith is there a way to tell wich one develops spots before engraving? Maybe there is something to lookout for like the converging lines or something.
hi can you explain how or where to set the output white point is?
His previous video on Slate engraving (The one with the Koala engraving on slate has an incredibly helpful section on this)
I'd only ask @mrcraftsmith if you would mind sharing all the testing files you've created? (The white to black range, DPI etc?)
I think this is the file you are after - let me know - tinyurl.com/digitalbaristagradients
Curious, what power is your diode?
Hey Michael - the Ortur LM3 is a 10w output laser
Great video and lots of info to take away from it. Is the file you have to get the white output value available to download at all?
Trying to get to a point where I'm happy with my image engraving and think this may help.
Hi Craig - thanks for watching - here is a link to the file - tinyurl.com/mrcraftsmithslatedither - if you haven't subscribed to the channel and you feel I'm offering some value to you - consider subscribing :)
@@mrcraftsmith thank you for the reply and I have subscribed.
The images that you prepared fully in Photoshop, when you opened them in Lightburn did you activate Pass-through or process the image again?
Thanks Craig - yes definitely activate pass-through as you have done all that needs to be done to create a dither if you process directly in Photoshop
@@mrcraftsmith thank you. Do you coat your slate with any products before or after engraving?
Hey Craig - yep I coat prior engraving with satin varnish
Are there any free softwares that can do this?
I couldn't be certain, but I suspect Gimp may have a levels adjustment similar to that shown in the video
Thanks for very useful info. Do you use something to finish the coaster surface? Shine or matt spray or wax or…? Trying to find something that would increase the contrast..
Thanks for watching and glad you found it helpful - for these examples I did sand the slate pre-engrave and coated with a light coat of satin varnish
#4 i think
I agree :)
4
Thanks for watching Merrill 🙂
Hello, do you definitively use photoshop or do you also use imag-r to obtain a photo to engraving?
I started out with imag-r but once I dug a bit deeper into Photoshop I have been using that - although I'm sure with more testing I would probably find the sweet spot with imag-r - I guess it comes down to whatever tools you have available to you and what feels gets you the best results - there's no right answer as to what is the best tool :)
I like the better contrast on number 5
Thanks for watching Ray