Sumea Modeller Challenge 9 Results!

wierd.. i think shika's is easily the best. what is dan miller thinking , giving it a 4?

Thats not the only weird comment/result from the judging.

This one specifically bothered me as it came from an industry rep.

"...Be careful with your use of tri’s and not quads. This causes a skinning nightmare..."

That's just plain incorrect.
In fact having triangles in a mesh has very little impact on skinning or weighting the mesh.
You skin vertice's, the amount of or distribution of edges that connect to a vertex play little to no part in how they deform once skinned, where as the vertice's relative position to the skinned joint(s) does.
Therefore polyflow and actual good skinning practices are the major areas to be aware of here, if the flow of vertice's is good for deformation then thats all that matters, whether it's constructed from a row of triangles or quads is irrelevant.

The engine will undoubtedly triangulate the mesh on export anyway. Nice misinformation. lol.

I know the judges are busy but they do have a responsibility to actually make accurate comments or else you'll help propagate bad work practices.

Anyway, nice job to all who entered, congratulations to the winners and I hope you all enjoy the prizes. :)

How Shika didn't win is beyond me, his entry was clearly the standout . (no disrespect to the other entrants)

seems like dan and sebasitan had odd choices for the best, maybe favouritism to their own students

I doubt there was any favouritism, what would they gain from it?

I will say inconsistency in the judging was noticeable though.

Apart from some strangely incorrect feedback, the main issue I noticed was judging seemed at times to be based on personal preference and less on the brief as a whole, if it was for a game asset then it would be judged on every aspect.

As an example only, Shika ended up with theses results: 5th, 1st, 9th, 1st, 3rd, 1st and 2nd ........ if all factors from the brief where taken into account you'd expect to see similar groupings in the resulting scores, yet we have some really unexpected scores.
Since the majority of the judges scored in a similar range you'd expect a final placing of around 1st - 3rd place, the 5th place is a bit of a stretch but could happen, however 9th place, well that suggests a problem in the process, in science you generally take these spikes as anomalies, you tend to throw them away as skewed data or redo the test.

While I used him as an example, I'm not saying Shika was ripped off, far from it, after all the five closest scores gave an average of 1.6 so 2nd place seems fair enough, I'm just trying to point out that you shouldn't see these wildly different scores from people if they are judging using the same set of rules.

Souri, maybe in future you could setup a more formal judging system? Maybe a weighted system which relates more to how an actual lead would critique work in the industry?
It would probably make it less work for the judges too, they would only need to score each section on a ramp based on it's quality and add a comment to validate the scores of each section.

I didn't intend this comment to be a swipe at the judges by any means, most people understand everyone involved would have taken there own time to do the judging and it is really appreciated, I just think as this challenge gets bigger it needs to become more fair and formalised, especially when companies are shelling out for prizes.

Although shika's model was a very original idea and looks very good, I think killa's model is alot better in terms of construction and textures and thats what the judges are looking at

are you kidding?

shikas model even though obviously not finished is brilliant...

leagues ahead of the rest

ya as you can see it can get quite subjective.. but this an interesting topic and something which annoys me about this industry .. especially when it comes to portfolios.. to many times a good portfolio can be turned down because of a personal preference.. In my opinion Shika's model is the best.. anyone who knows anything about character creation can see it would have been much more difficult to create shika's character.. yes the texture was simpler in layout than killadees but in the end .. it is about the end product.

Technically Killa Dees is solid. yes, but the end product is not as solid visually and creatively as shika's, I wouldn't put that character in my game but shika's ya , now that is something you would... maybe technically.. texture wise it isn't as crazy as killa dee's but then does it really have to be crazy? i would say simpler is better. One thing I noticed about Shika's is that it will Mip Map a lot better than killa dee's. All that detail in KillaDee's will get washed out to nothing unless you are close.. so then again this is just my opinion..

i think their should be a judges rating and a reader rating, just like gamespot and their voting for best game of the year etc.. I would say majority of people visiting sumea are connected to the industry in some way and some are more experienced than the judges who voted on this challenge. open the voting up to 50% judges, and the other 50% to readers of sumea.

It may be a nice addition for others to give kudos to the ones they like but I wouldn't have the final result dependent on user judging.

Opening it up to non industry judging would give even more skewed results, whether it's bias from friends, personal preference or just a bad understanding of whats needed in a games asset.

Industry judging within a formal structure is the only fair way to judge.
It gives the judges something to go by and the contestants would receive real world comments that a lead would give them if they worked in the industry.

This is after all about getting better at what you do and understanding the industry you want to work in.

i think the two guys from AIE probably need to get back in touch with the industry.. are they industry artists?

seems like the descrepencies wouldnt be so vast if they were seeing the same thing as everyone else ;)

Guys, there are even bigger discrepancies in votes for other challengers (look at the scores for Gogitisan, for example), but all this doesn't mean anything other than that judges don't have the same opinions. If you want to show the discrepancies in Shika's voting, then not jump on the 9 score in Killa Dee's results too? Because it's all pointless. This is just the nature of judging. Having more judges helps even it out, but the idea of having a checklist and point score for judges to mark and score against was done in previous challenges, and other than providing a tonne more work for the judges, it didn't really make the results that much better.

I have full confidence in all the judges I picked for this challenge, and I am completely happy with the results. If anyone thinks there's any form of favouritism going on, I would like to put forward that Shika is a pretty well-known AIE graduate too.

1. Anonymous Fri, 4 Jan 2008 01:43:30 EST

--------------------------------------------------------------------------------------------------
"...there are even bigger discrepancies in votes for other challengers (look at the scores for Gogitisan, for example)..."
--------------------------------------------------------------------------------------------------

That was my point, well that's what I tried to write. lol.

I don't for one minute think any reasonable minded individual here believes there is some clandestine plot going on but there were vast discrepancies which we wouldn't expect to see if judged within the same structure.

It had nothing to do with the judges themselves, I mean they did it for free, out of there own time, my issue was with the process that they had to judge in or lack there of.
The judges did perfectly as asked, I just think the judges should be asked a different set of questions. :)

--------------------------------------------------------------------------------------------------
".... but all this doesn't mean anything other than that judges don't have the same opinions."
--------------------------------------------------------------------------------------------------

You are right Souri, the judges don't have the same opinions.
Thats fine if you want the entries to be judged on which one was more arbitrarily pleasing to the judges, you'd expect differing results as they are not being viewed with the same critical eye, the work is being personalised by the viewer, much like a painting.
Unfortunately a single asset has more in common with the paintjob on your house than the Mona Lisa, it's only one of many practical assets that when used in context make art and should be judged accordingly.

I assumed, maybe incorrectly, that the entries were to be judged on there worth as a games asset via the brief you gave forth, if so then what is actually workable as a games asset should not see such drastic differences in the results, there would be some uniformity in what people in the industry expect.

There are some pretty consistent processes and limitations to go by in the industry and 1st and 9th place is not a reasonable difference when judged with them in mind.

One last thing, prizes are great but the ability to get real inside knowledge and critiques from industry professionals would for me be the only real reward to do this.
If the judges don't have the time to do a fair and accurate appraisal that's fine and quite understandable, they have lives and aren't being paid for this, however I for one find doing it just for a pat on the back and a prize all a bit narcissistic and pointless.

Now thats only my opinion, others may feel that the prizes and kudos are all they want, fair enough.

It's just I read your 'Sumea and 2007' post Souri and it seemed to me you wanted this site to become a place to help nurture talent and a place for industry types to become part of the process.

If so, Why not start with the challenges?

Anyway, as you can see I just love a good debate. :P

Heres hoping for a good year for the industry and Sumea in 2008 and congratulation's again to KillaDee, Shika and Thinkshake.

If you have any suggestions, I'm always all ears. Drop me a line in the contact me box outlining your solution.

i think dan giving shika a 9th place must be a mistake? can souri ask him the reason why it was given 9th? Dan has lost all credibility in my books if his going to make decisions like that? must be a mistake... if it wasn't then if i were you souri I would think about using other judges. If you don't think it is a problem , then do you really want this sort of debate everytime you have a challenge?

lets see some of his latest works.... ??