difference in accuracy (scanners)

[DMC]

The following is a pic of details of a few files we have to work with today.

On the left, we have files from a 3M ESPE scanner. On the right, we have files out of a Dental-Wings scanner. Now, look at the file size. See all of the 3M files are about three-four times the size? That's the amount of data the 3M collects. The difference is very obvious! The 3M scanners are the number one in accuracy hands down!! They collect waaaay more data points, with less voids and less smoothing and stitching. I have not collected any data from the new 3Shape scanner, but I have software to actually count the number of data points and the number of triangles. Anyone up for a challenge? Most of these files are single units, and a few larger files are bridges....

[/IMG]

 

[DMC]

Now is a pic of the teeth with the data points shown.

(on right)The Dental Wings has such poor resolution, you can see the points are too far apart for dentistry (more than 50um apart) (cough hunk of junk)

The 3M has soooo many data points that are so close together, the copings still look solid. (left)

[/IMG]

 

[charles007]

Start counting !! Like to know the differences in the Laserdenta and the new 700 3shape compared to the ST........

 

[DMC]

I'd be glad to do a shoot-out.

Waiting for any challengers......

email me your test case of any single unit.

I should have a master die we can all use.

Anyone wanna play?

 

[Scooter]

I'm impressed being a Sirona Inlab this really brings home some of my weaknesses in scanning information. Thanks for sharing!

 

[harmonylab]

well, it'd only be useful to compare scanners with same accuracy. I don't need to compare pixels to know that my inEos with its crappy 25micron accuracy is a lot worse than my etkon es1 and its 10micron accuracy. The difference in the scanned images is very noticeable. inLab MC XL's accuracy is 25 microns as well.. not sure if that's good or not... anyone know the accuracy of the etkon millers?

 

[alan2423]

you get a master die and i would love to scan it in my 3shape 700
and what file format are they saved in that can make a big differnece to the point cloulds, and points saved

 

[karabear]

I have definitely obtained the best results with the Lava scanner and software.

 

[Mountain Goat]

3M COS data points

Like ***'s last post about the amount of data captured in a scan....I need to find that answer for a 3M COS Digital Impression as compared to the Cadent iTero Digital Impression. Help me somebody!! I suppose I could call 3M and Cadent tech support, but I am sure someone on here knows the answer.......

 

[k2 Ceramic Studio]

Nice post, just looked at the raw stl data file in the everest scan pro for a single coping that we did this morn and it showed 5,701KB, could we not use a master stl die then have set peramiters, wall thickness margine thickness, and see what we all get back in terms of data captured in the coping stl manufacture.

 

[PDC]

The following is a pic of details of a few files we have to work with today.

On the left, we have files from a 3M ESPE scanner. On the right, we have files out of a Dental-Wings scanner. Now, look at the file size. See all of the 3M files are about three-four times the size? That's the amount of data the 3M collects. The difference is very obvious! The 3M scanners are the number one in accuracy hands down!! They collect waaaay more data points, with less voids and less smoothing and stitching. I have not collected any data from the new 3Shape scanner, but I have software to actually count the number of data points and the number of triangles. Anyone up for a challenge? Most of these files are single units, and a few larger files are bridges....

[/IMG]

Scott,

This is great info. I will have to make a decision in the near future on purchasing a scanner vs. upgrading to the new CS2 (Straumann). I am currently considering Delcam and 3shape. So if you get any info on the data points on those 2 it would be of great interest to me. Some forum members feel that the software is more important than the actual scanner...could you possibly clarify that argument? I would be in the camp of GIGO on that. What about u?

Maybe you could do a scanner comparison report for us...I'm sure the company who wins would pay you for your findings.

 

[Diadem Precision]

Hi guys,
I am humbled by the amount of knowledge, experience and expertise here. I may not have as much experience as most of the people here but for the last 8 years I have worked with some scanners [Nobel's Mod 50, Picolo, Forte, Lava ST, 3Shape, Dental Wings, Ovascan, Delcam] just to name a few. I did a lot of testing and comparison and I came to the conclusion that most of these are outputting a file that has more than enough cloud points in order to achieve a good fitting restoration. Now, of course the type of mill will make a difference since not all are the same. But, in my opinion the designing software is what makes the difference. Remember almost anyone can scan, and some of these systems will let you scan & design at the same time. When looking at the designing software you should be asking yourself - which one is the most efficient, user friendly and intuitive one? I still believe it's 3Shape even though others are starting to close the gap.
As for the quality of the scan - how much is enough? Most of these scanners have tolerances of 10-20 microns, don't you think that's enough considering the cement space starts at 50-60 microns?
Just my opinion - and I look forward to follow this closely and see some other people's thoughts.

Mitch

 

[Bobby Orr ceramics]

Cool info Scott. How would a doctor's Cerec scan file compare ??

 

[BobCDT]

Hi Mitch,
We also have about 6different scanners and CAD programs. All the scanners provide an end product that fits great under 10X mag. I am in total agreement that this should be all about the CAD software. The fit is there. Here is the real question. How quickly can you get to the best design? It's about mouse clicks. We believe it's 3shape. 35-40 clicks for a great full contour design. Less than 5minutes.
Bob

 

[DMC]

OK, so you may have a scanner with each single data point being +/- 25um ...BUT
Those data points are very far apart. The spacing is pretty obvious to me.
I can even measure some if you prefer.

What data are you missing in that empty abyss of space (250um-even 1000um!!)?

The 3M scanner has accuracy of around 7um, and the data points are around 25um apart. We are measuring two different things now. Data accuracy (how accurate is each single point?),and data density (distance between points).


You guys say 25um accuracy for a 3shape.....I say more like 250um. That's how far apart those data points really are, and those data points could even be wrong by up to 25um in any direction! End result......I don't need no stinking microscope to see the difference. You guys are either blind, or just kidding yourself.

 

[TheLabGuy]

Hi Mitch,
We also have about 6different scanners and CAD programs. All the scanners provide an end product that fits great under 10X mag. I am in total agreement that this should be all about the CAD software. The fit is there. Here is the real question. How quickly can you get to the best design? It's about mouse clicks. We believe it's 3shape. 35-40 clicks for a great full contour design. Less than 5minutes.
Bob

Designing only in 5 minutes? Origin might have 3Shape beat with Designing. I think the scanning part is what takes the longest with all the machines. 3Shape wins hands down, unless someone knows something faster that scans as accurate??????

 

[YMS96]

You guys say 25um accuracy for a 3shape.....I say more like 250um. That's how far apart those data points really are, and those data points could even be wrong by up to 25um in any direction! End result......I don't need no stinking microscope to see the difference. You guys are either blind, or just kidding yourself.

Prove it. Who cares how many points a scan has? If it works and fits, then that's all the matters. 3Shape has been proven to be one of the fastest there is and produces units that fit perfectly 99% of the time, so who cares about the points? It's all about the software!

 

[e...w...h]

Prove it. Who cares how many points a scan has? If it works and fits, then that's all the matters. 3Shape has been proven to be one of the fastest there is and produces units that fit perfectly 99% of the time, so who cares about the points? It's all about the software!

Absolutely right.

 

[DMC]

Prove it. Who cares how many points a scan has? If it works and fits, then that's all the matters. 3Shape has been proven to be one of the fastest there is and produces units that fit perfectly 99% of the time, so who cares about the points? It's all about the software!

How about I give you one data point that is dead on from a scan? Just one single point in space. OK.....now you have a 100% accurate scanner with piss-poor data density. Impossible to make anything.


but, but....my scanner is 100% accurate, how can this be?
Poor data density....just as important as data accuracy.


They go hand in hand.

Like trying to compare engines based on torq alone. No, you need to measure both torq and horspower.

 

[CHIVEINLAB]

I work with 3shape D640( with 2010 version 2.5.9.8 ) and single file

size is around 1,100KB and higher. small tooth #24 may less than

1,000KB but normar molar file size is 1,200KB or bigger