CG Gallery voting musings.

#41

Yes I would agree. The e-peeness factor is not overblown. The galleries are after all places to go to ‘showcase’ your stuff, and the lollipop stars from the viewing public can pump up the e-go, or deflate it.

#42

Hello!

Obviously nobody seems to take notice of my previous posting, but isn’t it possible to create an ‘if you want to vote you have post a comment’? I think that would be a solution to the problems without giving some people more and some less weight.

The vote alone is to uncommodating, but combined with a post there would be real critique. Even if it is a short one. And critique is what people are looking for. The good AND the bad! It would be just FAIR!

Besides: I am also against changing weights in advandage of ‘good’ artists. But I am glad the the discussion is back on.

cheers
Marc

#43

I don’t agree with oDDity’s assertion that award winners should have more weight in the voting process. Not all award winners make good critics by default and not all credible objective art critics are award winners here.
Though I think the overall voting atmosphere is skewed to the juvenile, this site is made up of a community of Cg artists at all skill levels (most likely more amateurs opposed to professionals). No one here should be denied or have diminished voting privileges on the basis of their skill level. In the end, and thankfully so, the moderators make the final, and more objective, decision on who gets awarded into the Choice gallery. IMO, they do a good job of it, most of the time, whether the voting mods are award holders or not. I’ve also seen a lot of artists here that do top notch work who are not in the Choice gallery, this does not disqualify them from making credible, experienced judgments on the quality of work submitted to the Showcase galleries.

Opeth brought up your question about voting privilege with comments required, back on page 1 with responses following.

As I mentioned previously, in my opinion, the reason why you see images with initially high star ratings (usually accumulated by a few initial votes) is because people will vote on the images they personally like and as a result vote high. Conversely people will not vote at all on images that simply do not appeal to them. Usually a work with very low ratings has much more than a couple of minor issues that need resolving, and would instead require lengthy global constructive comments which should be reserved for a WIP forum.

The stars start falling when mass attention is drawn to a clearly 2-3 star level work having a 4-5 star rating early on. That’s when the ‘adjustment stampede’ comes in. This is a natural trend and I don’t think it is malicious in nature most of the time. I applaud it.

Personally I only vote on images I like were I usually give 4+ stars (I can’t remember when or if I ever voted 5 stars). My votes are congratulatory in nature. Another thing to think about is the Showcase galleries (IMO) are there to show work not critique it, there are WIP/crit threads /forums dedicated for that purpose. To me, the star ratings simply indicate the level of appeal an image has to the voter. A low star rating does not mean the voter ‘hates’ the work or the artist. A low star rating does not mean the artist is entitled to a lengthy explanation by a voter as to why he/she got a lower than expected star rating. The star rating simply indicates the appeal level of the image. A comment would be nice but I don’t think it should be mandatory. I vote high without an explanation (comment), so why should I explain myself if I should vote low.

We are in the image business. If you get unexpectedly low ratings on an image submission just look around and compare your work to others. You can usually figure out what’s wrong with an image by comparing it to other similar more polished works, at least that’s an approach I use.

My biggest gripe with the Showcase galleries is the low bar of baseline work quality in what is accepted into the 2 and 3d galleries in the first place.

#44

In my experience, the best of any field are rarely the best suited to give advice. If you want to learn to play piano, going to a concert pianist is a waste of time, their job is to work on their own skills. You’re better off finding someone whose skills are in teaching others.

Keep in mind the CGChoice galleries are selected by the forum leaders, most of whom ARE in fact professionals in, or around the digital art industries.

But, yes, I agree with leigh, as one of the more respected artists of the community said ‘the stars feel like I’m back in high-school’. Personally I’m more an advocate of hiding any actual rating to public view and using it to present information more relevant to the particular viewer.

#45

Yep 100% right… some folks take that star rating way too seriously. I used to be one a looooong time ago, now I just don’t give a shit.:smiley:

#46

Oh I don’t know. Most of the CG tutorials and teaching materials out there are made by the great artists, not the great teachers. Seems in this business the 2 are one in the same. You just have those with time to teach and those without. Weighting a vote by achievement makes sense but ultimately I’d prefer an even vote for everyone. Much less whining overall.

Now this sounds good to me. How many times has your own opinion about let say a movie been swayed by what you read about it or its rating on this website. That’s why I try to avoid reading anything about a movie I want to see so that I won’t have an opinion going into it. Your opinion about artwork can similarly be swayed. I find that if everyone is saying how wonderful something is I will give it an extra critical eye just so I won’t be a ‘yes man’.

So you allow people the vote but hide all the results so that everyone casts a fresh vote. People that just ‘vote and go’ won’t read the crits anyway so they will likely not be swayed by them.

Then again if the votes are simply a way of grabbing CG Staff members attention then make it a very simple mechanism and hide all the results. But then I guess you would get all of the people asking, “Why?” Why didn’t I get selected, what was my score, blah, blah, blah.

I see why you guys shy away from this topic. :smiley:

#47

I don’t know why you all think the idea of letting experts have the only vote is such a bad idea. That’s how CG competitions are run, or entries for the ballistic books etc. You have a panel of expert judges, and they decide which is best, because they know what to look for.
I don’t know about you guys, but a nod of appreciation from someone who’s work I respect, means a lot more to me than 1000 joe publics’ going ‘awesome!’, simply because they gape at anything half decent.
The only problem is deciding who the best people are, and using the award gallery is just one way, another is to look at how many times someone’s portfolio has been favourited, so you know they have the respect of the general forum members. You could have a minimum requirement of 100 favourite votes in your portfolio before you are allowed to use the start system in the galleries, or whatever new method is devised.
My point is, that as you all agree, the current star system is pointless and you all ignore them anyway, however, if you knew the votes were coming from experts, or at least skilled artists, they would actually mean something and people would take notice of them again, because a lot fewer images actually would be getting stars.

#48

A lot of award winners are no better than those who don’t win awards anyway… so why should they carry more weight. Personally I’d favour the star rating system being disabled.

Cheers!

#49

The voting system, like any voting system is unfair in many ways and constantly abused. Mainly the big problems i see is that the already well known famous artists get their works rated high no matter if it’s great or if it sucks just because they are well known artists. Also artists who have friends helping them vote for them by rating their creations with 4-5 stars even if the work itself sucks, just because they are friends with the artist. This happens all the time and it’s definately not good for the artist as it corrupts the entire ratingsystem.

Lots of unknown artist, skilled one’s on this forum gets zero notice of their work, it quicky gets overshadowed by the famous artists works and the artists who’s friends rates them high because they are friends. I find this type of behaviour very disturbing, it’s way worse than people who rate you with 1-2 stars because they dislike you for whatever reason that may be.

Voting should be simple, only a thumbs up and thumbs down should be the choices when we vote on a piece of artwork on here. No counting of stars and such things as it corrupts peoples minds (and the forum) very easily.

/ Magnus

#50

It’s true that the award gallery is heavily weighted in favour of character art. If you just glance at the first page now you’ll that about 90% are characters, so anyone who models cars or spaceships or buildings is hardly represented at all.
However, in all 4 categories of modeling, texturing, shading and lighting, (not to mention animation) it’s much more technically difficult for a realistic character than doing the same for a building or car, so that a large reason why.
That’s why realistic human characters are more or less seen as the pinnacle of 3d achievement, just as paintings and sculptures of characters always have been through the centuries, and those threads can get hundreds of responses
You’ll never see a car render thread get that sort of attention, no matter how good or realistic it is.
Totally photorealistic renders of cars is something that was achieved years ago, while the highest limit for how well a character can be done in CG is some time away yet, and that also makes it more exciting in a way.

#51

Exactly! But before disabling it, I would try to change it and see if it works.

cheers
Marc

#52

I agree with that. Receiving a critique from a respected artist is always appreciated. The problem is translating that into a vote metric.

The forums at facepunchstudios.com have a fairly sophisticated method for measuring a persons merit in their posts and such and the community ultimately polices itself. A lot of coding would be involved with that interface though it doesn’t seem too complicated. Having used that forum though it seems like a necessary evil to control their often volatile user base. Certain parts of it are pretty good though.

Still my preference would be simple, uncomplicated with a slant toward positive reinforcement.

#53

Just to pick up on the original topic of the thread.
As an experiment I have been keeping an eye on my recent high profile portrait of a girl image.
It managed to stay at 5 stars for about 24 hours.
I don’t know if you guys are aware, but if you hover your mouse over the stars you get to see how many votes have been cast and what the average score is.
I have been keeping an eye on it and it has been gradually slipping down and is now at an average of 3.8, which proves beyond doubt that people are deliberately voting it 1,2 or 3 stars now, because that’s the only way the average could have fallen from 4.x down to 3.x.
Of course I have quite a few enemies on the forum due to my brash, opinionated manner, so that might explain some of it, but the other 5 or 6 images at the top of the award gallery page also have almost exactly the same score, 3.7/3.8, and I’ve been keeping an eye on the other recent entry by old-boy, and exactly the same pattern has emerged, it started off above 4 and has gradually dropped down to 3.8, because once people see an image getting accolades they immediately start voting it down.
I agree it’s not a huge issue, but it’s just rather sad and pathetic behaviour.

#54

That is exactly the trend I had noticed (not just for myself, it seems a rather widespread thing) and the reason why I started this thread in the first place.

I may add that I ‘caught’ one of the ‘one star voters’: I happened to check frequently on a newly posted work of mine, and, in a lull of the forums I got a sudden drop in average with only one vote added, the user name of the last visitor (the only one in a while) was still there. I don’t remember the name but I checked the profile: an old account (4 years or so) with no name, no location, no portfolio and not even a single forum post. I may be wrong but it felt rather purposefully anonymous to me…

I agree, it isn’t a big matter, but it’s irritating, I’d rather prefer a ‘thumbs up or nothing’ option, if a voting system is needed at all, that is.

#55

it could work if you want to give a star, you have to write comment. than you are able to add your vote, what will be visible beside your name.
than maybe you will have at list an explanation

#56

Does it? How do you know? I thought the same in the first place about this troll stuff, does it really happen on a big enough scale?
Things do tend to go down but as there is nothing to vote higher than 5, its likely to go down and very hard to stay at 5 on average.

Anyway the digg way seems like a good idea, as people have said it does work elsewhere. No need to vote down, just, I likes, or don’t touch it. But I don’t really see much of a problem anyway just interested reading these threads.

#57

Here’s an example of what I was talking about earlier. Rod’s piece is very good and I left comments but for some reason I didn’t vote. Nothing to do with the piece I just hardly ever think about voting. The piece has over 44,000 views but only 224 votes so I’m not the only one obviously. I’ve no idea how to fix that problem but maybe the Up or Nothing vote would help. And if it was that simple you could append the vote icon to each post providing a better reminder.

BTW, since you mentioned it which reminded me, I did go ahead and give you 5 Rod. :smiley:

#58

As I mentioned twice already, the rating will always be the highest in the beginning due to the fan-boy rush. If you are blown away by an image you are more likely to vote on it and vote high. If you are impartial you are less likely to vote at all or vote lower, if you do. The rational (more objective) votes seem to trickle in over time and are represented via a large vote count. The average rating in the 3d Choice gallery is around the 3.80 mark, since I can’t average the averages I can’t give you a precise figure.

More ’musement:
If the star system were based on the actual value range of the star, instead of rounding up to the nearest value at the halfway point, only 16 images in the entire 3d Choice gallery (to date) accrued 4 stars (4.0~4.99), just under 2% of the 834 counted. The highest rated image in the 3d Choice gallery was this one at 4.4455 stars with 770 votes. Getting anywhere near that will be a long stretch for anyone. It’s rare enough for an image to get from 3.8 to 4.0.

So oDDity, your image not maintaining a 4.5+ average in the long run is more par for the course than evil in the system.
For the record I gave you 4 stars. I would have given you 4.5 but the system doesn’t break down that way.

#59

I don’t think the problem is inherent in the system, but in the voters.
Human failure is always the weak link of any chain :smiley:

#60

What’s interesting is that the average score in the 2d award gallery is more like 4.2 as opposed to the 3.8 in the 3d award gallery. There is a large percentage of 3d images with only 3 stars, while a few 2d images have even managed to maintain 5 stars.
Make of that what you will.