Voting issues

[gohkm]

I resisted posting my stuff up for years on CMON. When I did, I wasn't even invested in the voting system or my ranking - it's not why I paint anyway. In the end, I found that Facebook was a much more positive environment in general for posting pictures of my models, and the environment was more controlled - out-of-control commentary is easily removed so focus stays on discussion of the model.

Admittedly, from what I read of the Discussions and WIP sections here, the feedback is more directed and constructive, but I think for me, Facebook is the way to go. I simply cannot be bothered posting in here any more.

Besides, it seems that CMON has a 'preferred' painting style. Going out of the box to experiment doesn't seem to do very well at all.

 

[Darthmarsh]

I must say that my experience with CMON has been 100% positive so far (but then I haven't posted anything into the gallery yet). I lurked for ages admiring all of the different projects and paintings. When I finally did start a WIP I found the community, advice and support incredibly helpful.

There are lots of good ideas here on how to change the system to help make it better but we could all do one very simple thing to help without a huge overhaul. Vote on the gallery more, even if it is a few mini's a day. and I don't mean on just the obvious one's. Lets all make a conscious effort to vote and comment on the not so good ones as well. vote in plain sight but then offer constructive advice on how to improve. Change starts with us people. If we make the effort then maybe more will to

Edit: voted and commented Webmonkey

 

[cassar]

I think the top last 7 days should be optional ie you should be contacted prior to the image being posted there, I know I'm being a point whore but I was more than happy with my image scores then if they get to the past seven days gallery its like "targets will fall when hit" namely a snipers paradise I personally would prefer an opt out option oh and why not just lower the voting scores from 50 to say 30.

 

[BloodASmedium]

I AGREE WITH CASSAR . I said to myself this, why so many of my top work makes the "new and notable" only to have the scores dropped 7-10 full points . The funny thing is only when you make it there do you get wiped out. So for awhile I was scared: scared because 90% of my works make the top last 7 days. But then I said this.. Instead if me crying out at something so uncontrollable think of it this way. If my painting became soooo good and survived the dreadede " NEW AND NOTABLE". Then I could consider myself a great artist. So the trick. Well it's easy - practice practice practice until your model painting is sooo damn good I mean good enough where even 10 snipers cannot undo what the 100ds are saying. Again soo good that for every sniper who can't stand those better then him/herself, that there are 50 or so others who hail your work. So again it's easy keep taking feedback , hone your skills with research, questions , literature and the seminars by top artists, and your own w I p , so that it's sooo damn good so that nits fruitless for the small amount of sniping to make an effect. Simple.. If you survive the top "last 7" zone then consider yourself an accomplished painter. Hang in there it happens to us all!

 

[Century]

The few times I've ended up in the top in the past 7 days / new and notable, I don't think my scored ever changed by more than 0.1 or 0.2.

Now, the following may sound harsh, and I hope no one takes offense to it, but if your model is dropping that much, perhaps it's going to it's "normal" value? Like I've said, I've never had one drop that much, but I also never link my submissions anywhere else (other forums, blogs, facebook, etc.), so I don't get a "friend" bump. If you hit the top in the past 7 days with more than about 40 votes, you're model probably isn't following the "normal" trend (from my experience), so it's score may not be indicative of a more varied audience. That said, this certainly isn't the case for every model, and snipers can and do make large effects on the score, this just hasn't been my experience.

I watch the voting on my minis pretty closely once I post them and the trend I notice is this:

Higher marks initially
a drop when entering in the 20-30 vote range
rise back when around 35-40 votes
If I manage to make the new and notable that score may drop a bit, but the change usually isn't huge.
The ratio of non-registered to registered voters varies from about 3:2 to 2:3 (I usually have more registered than non-registered votes, but only by a small amount, with most being about even)
The largest difference between the registered and non-registered score is 0.4 (50% where registered is lower, 15% where registered is higher, the rest (35%) are the same)

This is all my own experience though, so I'm not sure if what I've seen is typical or not.

 

[ten ball]

thought I would get over the snipers & trolls as time went on but if anything ive got worse and more anoyed about it all.
In the scale of things in life its a small problem but of late it's got me so angry that I won't be posting anymore Picts to my gallery, will be concentrating on my WIP and articles. If I do post a final project it will be to putty & Paint where sniping is not possible.

 

[BloodASmedium]

Well I really do t know what the deal is. What I know is that the moment I've started a w I p along with those others that do the scoring and amount of votes was not the norm . It was much higher from things like more people were exposed to my work as others who have done the w I p logs. Try doing a w I p GOHKM and see if that helps . I do know its all random so it's really a vote of possibly 9,000 members. TO ALL READING THIS TODD SWANSOM IS RANKED 600 and change. NOT EVEN IN THE TOP 10. FUNNY THING IS HE HAS 8 count Em 8 (GOLDEN DEMIN SLAYER SWORDS) AS WELL AS 13 GOLD 1st PLACE GOLDEN DEMIN STATUETTES. I DINT CARE WHAT RATIOS ANYBODY SAYS. THE SITE REALLY ISN'T A SYMBOL OF WHO OR WHAT THE ACTUAL VOTES ARE. SO NOBODY CAN TELL ME ANYTHING THE GR8TEST PAINTER IN THE WORLD IS TODD SWANSON . SO THERS THE EVIDENCE OF PEOPLE BEING HATERS!! Bam right back at ya!! Lol

 

[BloodASmedium]

AGREE WITH TEN BALL AND ALL THE OTHERS THAT SNIPING IS A CHILDISH WAY AND DOES AFFECT THE INTEGRITY OF WORK DONE INCLUDING MINE . AND THE INTEGRITY OF THE CMON SITE IS AFFECTED AS WELL. Sorry century I'm glad it's never happened to you bud. It kinda hurts and is infuriating when it does. There 4 people me ,GOHKM , ten ball and many other high profile artists that it happens to very steadily. It's really sad and were not making it up bud.

 

[Century]

It's really sad and were not making it up bud.

Never said you were making it up, just that there may be reasons outside of snipers, of which I proposed one possibility. I've had models sniped myself, just not noticeably after reaching the new and notable.

 

[BloodASmedium]

And again I def wasn't convicting you I should've explained myself better. There are I'm in agreement many many possibilities. And as I mentioned its all a problem to real for all who work really hard and spend on the new and notable as a way for picking up business , contracts and the like. It's a problem cause the one possibility if the sniping can affect income to many of us who do this as a full time gig. There are us who use it as recognition so when we do seminars alot of the forum public know who we are and helps in catering to how you g artists aspiring come to know and trust out work through that last 7 days and the new and notable. Put it this way it's more of a problem as a possibility than it not being a problem. Again this is coming from high ranking cmoners such as myself and andyg ten ball , avelorn Etc. merely saying as a possibility it can do more potential harm than not. Hope this clears things up.

 

[MAXXxxx]

...that it happens to very steadily.

someone must really hate lions The lioness bust went down then back (so sniped and autocorrected) 3 times this week. Was funny to see number of votes go up (and the score down) and back again.

A bit similar happened to Kitty too, but I think it was some bot going through the site or snipers getting clever.
Instead of 1-s it got 6 6-s relatively rapidly (within an hour), dropping the score by about 0.4, and as they aren't too low scores they won't be corrected. (and quite frankly the score got to a more believable level too)
On the plus side the 6 votes no matter how 'low' pushed the no. of votes over 50 :lol:


the only thing that bothers me with all these sniping and such is: "ok you don't like it at all, and you give a 1 for that. Fine, but WHAT was that bad in your opinion, so I could work on it on the next mini.". In other words I'd gladly accept even the 1-s if there would be a comment saying this and this was bad and that's the reason.

 

[chaos spawn]

I received some 3s on a mini, but most of the time snipers give me 1s which gets corrected after a while.

I think the snipers sometimes have a large impact on the score. When I vote I sometimes don't really know which score I should give. Then I look at the current rating and try to figure out if that rating is reasonable or not. My vote is usually not more than 1 point away from the average, sometimes higher, sometimes lower. I think a sniper that lowers a score can affect following votes of other members, too.

But I continue posting my stuff here anyway, because I like having everything in one gallery. If someone is interested in my projects I just need to give him one link and he can look at all my recent works.

I think you should not be repelled by a few snipers. Most of the people here take it seriously and are very fair. Those snipers should not spoil everything. After all voting can never be perfect. People have different tastes and some minis will always be underrated while others are overrated.

 

[gohkm]

It's not actually the snipers that bother me; in any case, CMON runs a correction algorithm that normalizes scores about 8pm Sydney time every evening.

What irks me is that CMON is largely a popularity contest. The more popular artists not just get more votes, they also have what seems like greater leeway in experimentation. The vast majority of CMON voters seen to prefer very clean paintjob, or those with weathering effects done up to a really high standard.

Anything else, and you automatically score low. And, as it has already been pointed out, often, there's no commentary other than 'get better pics' - not exactly precise criticism, since image quality may vary wildly depending on device used.

I don't get this when I post on Facebook, the tone is generally more positive, but criticism delivered can still be brutally frank. I'll start a WIP thread when I return from overseas, so I can gaug and the difference between Facebook and here.

 

[Pictish Mini Painting]

Oof, just looked into that Putty and Paint site.

Some lovely models there, and a lot of things I saw that I would be interested in sourcing to try myself, some nice busts too which is something I haven't tried before.

Don't think I am good enough to put stuff near that yet mainly as my photography skills still suck and make my stuff look awful.

But great idea generator and skill pusher.

regarding the voting on here, yeah it takes a while, usually peaks at about 30 for ages then trickles its way to 50+. I made a very similar post to yours 'nels0nmac, about a year ago and its just the sheer number of stuff to see on here now and a small pool of voters it just takes time

 

[Avelorn]

Having access to the distribution of the scores on individual submissions I can say that sniping is much much less of a problem in skewing the scores than friends or followers being "nice". Easy to see for people who are cmon supporters in that the score from the registered votes (one per user) is most often lower then the combined score, that also counts anonymous voting.

 

[MAXXxxx]

I had too much time at work... so I wrote a CMON votemachine

It won't help a lot (as I don't really plan on releasing it), but in time(it's not that fast) it'll give an extra vote to each image I haven't voted for yet.
For scores it uses a really simple algorithm:
- take the ceiling of the current average. (supportive, even a 6.1 is rounded up to 7)
- if registered score differs by 0.7 compared to the original score (before the rounding) adjust it +-1 as needed
- take care not to give a 11 or 0
- done.

the best part is that it can run in the background during work.

And why did I use the algorithm like that? because I noticed, that I don't really differ from given scores at all even if voting by hand, so it doesn't behave too differently than I am.


edit: I have to take it back. Apparently the list for "yet to vote" gives back those where I voted already --> not usable feature.
So I have to figure out how could I get a list of images where I haven't voted yet.