On checking with the latest build 856, the Match surface is actually producing worse results than before.

You do not trim back a surface before using match surface.

Match surface is used to match (G1/G2) 2 touching surfaces, not to loft a gap.
In the case where a loft has been made, then that should produce a G1 surface, however, at times a G2 is required, so you would then use match surface on the loft/connecting surface, which does not work in VC/Shark.

Probably the reason Tim has not taken time to fix this tool, it looks like very few actually know how it should work.

Its easier to show than explain:-

See attached mov.

I have created 2 surfaces (similar to what is shown in the VC user manual for match surface). I make G2 match surface first in VC pro which is incorrect result, I then perform G2 match surface in Rhino (on same geometry) with expected result.

http://dc179.4shared.com/downlo...4dd066/match_surface.mov

Needs fixing as a priority !!!

Unfortunately it does appear to be the way with this product, that tools are party implemented, then they can be added to the list of "functions" to make it look good. But using it is another matter.

As Tim stated he "Will check this out shortly.." and then went on to "but I understand what's going wrong and need to set aside a couple of days to fix. It's definitely got my attention.", that being 18 months ago, I do start to doubt his ability to correctly implement this tool.

I have to agree, this imho needs to be on the very top of the priority list.
It needed to be fixed ages ago.

At least in Version 5.5 there was some functionality, although quite limited.
In Version 7 it just stopped working alltogether.

Please have a look into surface matching in Rhino, where for instance position-, tangential- or curvature matching for adjacent edges can be chosen seperately when creating a net surface.
If you want to obtain a matched surface like this in Shark 5.5. you would have to match every edge one after the other, oftentimes leading to bumpy uneven surfaces at the end of the process (matching matched surfaces doesn't really work that well).
If You want to try to obtain this match in Shark 7, well.... just forget about it.

Instead of improving functionality and fixing bugs, we are moving backwards in an area where Shark/CU already was quite weak.

I was going to upgrade my package to get the surface modeling tools, but if they still need work, then I guess it's best to wait.

Mike...

Tim, I think an update on the status of fixing the issues of surface matching would be appreciated.
It is not a good sign when users start to use older versions or other software (rhino) to do basic surface modeling operations that are bearly usable or broken in Shark 7 (that user would be me, btw.)

I don't expect a software to be perfect, but when there is a serious problem, I'd like to know that the developers are aware of it and working hard to resolve it.

Just to put things into perspective:
this thread is over 2 years (!) old, and the only thing that changed in this timeframe is that a tool that had users complain about its functionality stopped working totally.

I think we will be hearing from Tim soon regards this issue, there are many of us who really want shark to be the modeller of choice - it has some great features and I really like the interface ( clean/ intuiative / mac freindly ) . I am also looking at rhino - but the interface is a turn off !!! lets keep plugging away on this issue as I am sure it is being looked at - Tim ?????

I've watched the video Steve posted, and I watched the youtube example referred to. In Steve's video, there are two similar surfaces. When Shark's match surface is applied, it makes a hash of it. When he uses Rhino's tool, it achieves the desired result, but I can't see if there's user parametrisation of it. Steve used a ruled surface for his example, the tool arbitrarily created a compound-curved surface to match to it, with only the opposite edge remaining of the original. Does Rhino allow parametrisation of the surface? Can you specify the blend distance into the second surface? It looks to me as if the tool notionally creates a continuous blend entirely on the second surface, with zero offset into the first surface, is my understanding correct?