New EYEMAGIC Scanners EMS7000

[Solarfire]

Firstly let me make it clear, this is not an attack on anyone or taking sides rather an analysis of the ILDA pattern.

It was said above, that Solar's ILDA pattern was pretty much as good as it gets.

I'd dispute that you can't get a better ILDA pattern because this is what I get from the Kvant side by side with Solars.

Please excuse:

1. The single colour - I have no red and the blue was making it too bright for the camera to record

2. The quality of the photo as I only have a compact camera to play with and thus I had to over expose it and compensate for exposure in photoshop - hence the fuzzy lines and the dots that appear as lines instead of dots. I can only assure you that they were dots on the wall and the lines were razor sharp, pencil thin and perfect.

Anyway, here it, this to me is pretty much a perfect ILDA pattern - drawn with LM scanners at 28K angle unrecorded but scanner speed adjusted to compensate to ensure the circle sat in the box as it was supposed to:




One thing that shows up straight away despite the camera's exposure "fuzz", perfect striaght lines, corners, the dots are straight (albeit over exposure makes them appear as lines or elipses in the case of the lower dots, in reality they were perfect dots on the wall).

It was said this was the best @ 62kpps @ 10.5 degrees, would you mind dialing that in on your scanners and making another picture, then the comparison will be apples with apples...

[Solarfire]

@ Pangolin

Thanks for the insight, now that I understand that it wasn't even engineered explains a lot in regards to the lack of professionalism in its design and lack of bandwidth.

[norty303]

Thanks for the insight, now that I understand that it wasn't even engineered explains a lot in regards to the lack of professionalism in its design and lack of bandwidth.

That said, it still tests the scanners in a number of important ways, and provides a standardized benchmark by which you can say one scanner performs better than another.

A scanner can either scan X number of kpps at Y degrees or it can't. One that can, is likely to be a 'better' performing product than the one that can't (at least at the extremes of speed, which is ultimately what we're looking for, I think).

To go back to the car analogy, we seem to be getting into the realms of saying that the speed gun is imperfect/inaccurate, therefore the results are not valid.
So long as we are all using the same speed gun, it shouldn't matter.

[Solarfire]

That said, it still tests the scanners in a number of important ways, and provides a standardized benchmark by which you can say one scanner performs better than another.

A scanner can either scan X number of kpps at Y degrees or it can't. One that can, is likely to be a 'better' performing product than the one that can't (at least at the extremes of speed, which is ultimately what we're looking for, I think).

I agreed norty, if your main objective is reproducing the ILDA frame but not in interest of a broader balance when reproducing actually show material which is filled with more dynamics then the static content of the ILDA frame.

[norty303]

I think we'll have to disagree there.
There ILDA frame demonstrates ability to start and stop quickly, sharp corners, ballistic curves, etc
I would think that it is entirely indicative of the sort of thing you would want to repproduce in show content.
Tuning for specific types of show content may well satisfy that type of show content, at the expense of other content key attributes.
That's all well in good in very narrow applications, but doesn't give good balance and versatility. 'Jack of all trades and master of none' is probably where 'most' of us want our scanners to be (imo)

[p1t8ull]

I would like to see some graphics and raster shows sent down a splitter to both EMS7k and CT6215H and the speed gradually ramped up for a real world test. If the EMS prove to show sharper corners etc then I can see no reason why they can't be described as the fastest scanners on earth.
As already pointed out, Eye Magic didn't say what they were fastest at..

The 'Creation' show has always been a good 'Nurburgring' to show up a good or bad set of galvos, imho.

[norty303]

Dsli Jon posted a comment on a thread (maybe even this one - I'm buggered if I could dig it up!) about his subjective differences between CT and EMS scanners (the 4K's I think) and why he preferred one over the other for certain work, which worked on their respective strengths. He used words like 'hard' and 'soft' (iirc) to describe the graphics they produce, which may well relate to how effectively they can render those image characteristics

[White-Light]

It was said this was the best @ 62kpps @ 10.5 degrees, would you mind dialing that in on your scanners and making another picture, then the comparison will be apples with apples...

My scanners are at their rated speed and that's the point and that's why it is apples for apples.

Extra speed doesn't mean you can have a less than perfect pattern. If properly tuned and capable of doing X speed the pattern should be perfect at X speed.

[White-Light]

I agreed norty, if your main objective is reproducing the ILDA frame but not in interest of a broader balance when reproducing actually show material which is filled with more dynamics then the static content of the ILDA frame.

I think we'll have to disagree there.
There ILDA frame demonstrates ability to start and stop quickly, sharp corners, ballistic curves, etc
I would think that it is entirely indicative of the sort of thing you would want to repproduce in show content.
Tuning for specific types of show content may well satisfy that type of show content, at the expense of other content key attributes.
That's all well in good in very narrow applications, but doesn't give good balance and versatility. 'Jack of all trades and master of none' is probably where 'most' of us want our scanners to be (imo)

Exactly. Norty, I can't be bothered to get the laser back out now, but if did then I could show exactly how the ILDA pattern relates to real world show performance.

If I take my scanners up to 30K a mild amount over speed, then the small circles in the colour test just start to show imperfections ie they don't join perfectly. If I then go to some show frames with circles in both graphics and beams then I find exactly the same result in the real world cue display, the circles don't join exactly. That is why its still a valid test.

[mixedgas]

Tom's taking a big chance with the "Y" cable test.

One thing, we all adapt the number of blanking and corner points and anchor with sliders in our software, to a given scan pair. Tuning of the amplifier matters, so does software settings, on these more subjective tests. This is why bench testing, with instruments, is also important.

Bruce Rohr, of Cambridge fame, published a nine page paper on how to standardize galvo testing, but I have been unable to obtain it.

Steve