pouët.net

이보크 / evoke 2019 | August 16–18 | Cologne Germany

category: parties [glöplog]
Quote:
Regarding voting: I was also quite surprised about the top two entries in the tracked music compo but what do I know...


Same happened to me, looking at some of the results of the ST-compo in 2017 and well, it´s always unpredictable how you end up at parties.. unless you bring a football team of group mates. ;)

However, even if someone might find my idea totally boring, but why not bring a -regular- tracked music compo to Evoke. Maybe even as addition to the "special" tracked stuff. To be honest, the ST-01 compo was fun, but the last two years my attention to do something for the tracked compo was zero.

I think quite alot of musicians would like to contribute with a "normal" tracked music entry at such a huge event - maybe just give it a thought for next year, guys.

Hopefully see you next year in Cologne!
added on the 2019-08-21 10:19:32 by Dascon Dascon
The thing with the "special" music compos came up years ago, during a period in the scene where tracked music compos were almost dead, or were "unofficial" at best. It was the aim to give new life and purpose to tracked music compos. Now, the situation is indeed a bit different, and there are far more active tracker musicians around again than there were ~15 years ago.

Still, a "normal" tracked music compo would be a bit dull. Maybe we'll bring back the "tiny music compo" from the 2000s (modules under 64k). Or maybe something completely different. But yeah, tracked remix compos have kind of ran its course by now.
added on the 2019-08-21 12:53:49 by dipswitch dipswitch
Thanks dipswitch, in fact I missed that part as it was probably around my time of inactivity (I stopped scene activities around 2000 something and returned around 12-13 days later).

The "tiny music compo" sounds amazing, though. Let´s go down to max. 10kb. ;)
added on the 2019-08-21 13:04:15 by Dascon Dascon
Quote:
(I stopped scene activities around 2000 something and returned around 12-13 days later)


LOL.. I mean YEARS.
added on the 2019-08-21 13:05:05 by Dascon Dascon
my 2 cents: the votes should be investigated - at least for 4k and alternative platforms the results are completely bonkers and unacceptable IMO....
added on the 2019-08-21 13:16:56 by oerg866 oerg866
Quote:
Maybe we'll bring back the "tiny music compo" from the 2000s (modules under 64k)
Oh! That sounds like fun <3
added on the 2019-08-21 13:30:02 by numtek numtek
Hello fellow demosceners,

we picked up on the discussion about the results of the voting in
the compos of this years evoke and would like to let you know that
we investigated if something went wrong. Because we care about fair
competitions a lot.

TLDR first: we do not think there was something fishy going on.
we don't think there was (massive) namevoting and we just basically
found out that the competitions were really close in votes and
about ten more people voting could have made the difference between 5th
and 2nd place. There _was_ a really short "peak" in network traffic
during the last entry of the pc4k compo, nothing the server couldn't handle afawk,
but this might be a reason for "Eisenerz" having less time to be "livevoted".

As an example of what we did we used the pc 4k compo, since that
seemed to be the "weirdest" result. All the stuff that we analyzed
also possibly counts for the other competitions at evoke.

Due to interface/technical reasons, when running a competition,
we have to enable Live voting for every single entry, we fucked up and did
this way too late for "Eisenerz". We also talked about "process-problems"
that might exist with livevoting being closed directly after the compo and hence maybe
penalizing enties that are played later in the compo.
This reduced the time to 'livevote' for "Eisenerz" and might have influenced
the outcome of the compo.

Data suggests in that direction and people at Revision leave the
livevoting up for five minutes after the compo, we didn't, we will in the future.

We are really sorry this happened and promise that we are working on fixing this
problem and reduce the chance of human errors.

To be completely transparent and also to ask you to find further insights we
would like to give u some insights into what we looked...maybe you draw
different conclusions.

Here is what we did:

Please keep in mind this is two days after the party and in general
just a "post-mortem" look at things, so numbers might be off due to
sql statements that are not _exactly_ what partymeister does, the gist
of it should be clear though:

general data:
Quote:
visitors at the partyplace: around 460 (plus around 20 organizers)
used votekeys: 318
votekeys with at least one vote: 285

pc4k compo votes at 01:00 sunday:
Quote:
472 Druckbestampfung
417 Glitch Rider
408 One Way Trip
394 완노비
364 Precious Confusion
305 Eisenerz
303 Lawnmowering up the angels' skirt
276 DØDELFRUIT
215 PiaggioNonTroppo
213 R0bbensuppe
191 Bitty Twister
169 작은 (small)
165 nop/e.raw
159 Best Evoke

results of the pc4k with number of votes and vote average:
Quote:
Glitch Rider 630 147 4.2857
One Way Trip 591 145 4.0759
완노비 555 135 4.1111
Druckbestampfung 555 143 3.8811
Eisenerz 538 117 4.5983
Precious Confusion 525 132 3.9773
Lawnmowering up the angels' skirt 406 127 3.1969
DØDELFRUIT 372 112 3.3214
R0bbensuppe 313 109 2.8716
PiaggioNonTroppo 298 111 2.6847
Bitty Twister 264 103 2.5631
작은 (small) 241 105 2.2952
nop/e.raw 232 98 2.3673
Best Evoke 208 92 2.2609

network traffic during the pc4k compo:
BB Image

unique ip's doing "POST"-requests to partymeister during the pc4k (to rule out DoS/DDoS):
BB Image

Discussions about voting results are most likely as old as the demoscene
itself. It’s unfortunate that all of this happened and you can be assured that we will
be taking measures to prevent our mistakes from happening again. However, as we can see from the data,
it is something you can also help us by actively voting during the compos and afterwards.

on behalf of the evoke organizing team,
steam
added on the 2019-08-21 17:13:15 by steam steam
Thanks @all for this awesome party!

I'm very happy about everyone who enjoyed my music in "connect me". I was a bit sad on saturday that I didn't manage to get through preselection in MP3 compo, but the positive feedback to my SID made it up for me :-) Especially seeing some people dancing in the tent @ the compo ... <3

Tbh I expected a lot more of the "annoying music" feedback as I'm aware that the composition can be quite "challenging" for bleep-blop-in-your-face untrained ears ;-)
added on the 2019-08-21 18:09:26 by spiderj spiderj
@spiderj - sorry for that, but we had 29 entries in the compo and had to cut it down to 15-16. however, unlike some other parties, we have a preselection procedure which is as fair as possible to all participants. read more about it here.
added on the 2019-08-21 18:41:32 by dipswitch dipswitch
@dipswitch: I know. And I have zero problems with that... Being a little sad for not making it into the compo doesn't mean that I think everyone has to like my little shit-chip-tunes ;-) I'll keep it for some other party...
added on the 2019-08-21 19:26:05 by spiderj spiderj
@dipswitch: that’s really a great way to preselect, very good job you guys are doing there!
added on the 2019-08-21 19:32:46 by Dascon Dascon
Quote:
as I'm aware that the composition can be quite "challenging" for bleep-blop-in-your-face untrained ears ;-)


My ears have been very well trained for "bleeps" and "blops" coming from PAULA and SID during the last (almost) 3 decades. So that sure is not the reason why I fnd the music annoying. :)
added on the 2019-08-21 19:37:24 by StingRay StingRay
Quote:
My ears have been very well trained for "bleeps" and "blops" coming from PAULA and SID during the last (almost) 3 decades. So that sure is not the reason why I fnd the music annoying. :)

Guess "in-your-face" was the important part. I already know that you don't have problems with SID per se from your nice comment about my music in the last AC demo from Nordlicht ;-)
added on the 2019-08-21 19:44:42 by spiderj spiderj
Since the discussion is already on: Making sure that all entries, most importantly the last one, get a fair amount of time to vote on during live voting is important. Anything that helps automating this would help avoiding people making mistakes will help.
At this point that might mean that PartyMeister should handle all of that automatically, e.g. by automatically closing the compo x minutes after the "end of compo" slide was shown or automatically opening voting for a specific entry if it was shown for more than x seconds.

I know all of that has its own share of issues (in particular if orgas navigate to the wrong slides without realizing) but it might still be better than the current situation which heavily relies on people hitting the right buttons at the right time (not too late for the entries, and not too early for closing the voting).
Evoke 2019 was another great party; it was lovely to celebrate with everyone else there too and enjoy the compos, DJ sets and other events which occurred within the main hall.

This was also, according to my wife and myself in particular - the friendliest Evoke party yet, where quite a lot of sceners wanted to talk to us. It's always nice to catch up with people such as Havoc, Lug00ber, Shifter & Honkey, Gasman, FraNKy, Metoikos, Truck and Topy44 amongst many others - and it was also fantastic to finally put a name to a face (and he's a king for sure within the demoscene) - this was BeRo of Farbrausch, who was sitting opposite us for the duration of the party and was also sharing our electric point, which we didn't mind at all.

The party weekend went very smoothly. We had driven over from the UK and our journey to and from the party was also relatively smooth, apart from some delays on the home journey via the Eurotunnel where there had been a train breakdown but this fortunately did not go on for too long.

We will be back for the next Evoke (B-word or not) so we look forward to seeing you all then.

Many thanks to all of the organizer team. ! We had a serious blast :)

The selection available at the bar appeared to have improved quite a bit this year - even though there had been no problems before. Loved the mini hot dogs, please have these available again if possible. Service was fantastic here too.
added on the 2019-08-21 20:49:23 by Felice Felice
@Felice, the mini hotdogs were there for the past couple of years, so I guess they're here to stay! =) They are indeed an excellent inbetween snack.
added on the 2019-08-22 12:20:35 by dipswitch dipswitch
Quote:
This reduced the time to 'livevote' for "Eisenerz" and might have influenced the outcome of the compo.

It "might"? Knowing that live voting was enabled "way too late" for "Eisenerz" and closed right after the compo, that gives "Eisenerz" which is 2:32m long a live-voting window of maybe 1 minute. Judging by the crowd reaction, the live-voting data provided and the average votes I'd argue that "Eisenerz" achieving somewhere between 400-450 points during the live voting instead of the 305 it got within maybe a minute of being voteable is a way more likely scenario than what you suggest, in my mind there's not really any question about it and it was not at all "really close". I love you guys and shit happens but please keep it real and don't downplay the effects it most certainly had.
added on the 2019-08-22 13:19:00 by LJ LJ
Quote:
Due to interface/technical reasons, when running a competition,
we have to enable Live voting for every single entry, we fucked up and did
this way too late for "Eisenerz".

Quote:
This reduced the time to 'livevote' for "Eisenerz" and might have influenced
the outcome of the compo.

Data suggests in that direction and people at Revision leave the
livevoting up for five minutes after the compo, we didn't, we will in the future.

Quote:
We are really sorry this happened and promise that we are working on fixing this
problem and reduce the chance of human errors.


this is as "real" as it get's i think. we just will never know for sure IF people would have voted if they could have, and how, and how many. hence the "might". again, "data suggests in that direction".

we are not trying to "downplay" our fuckup at all and we are sorry if you read our statement that way.
added on the 2019-08-22 13:33:27 by steam steam
This might've been true abouth te PC 4k compo, but what about the other compos with contested results (alternative platforms, PC demo, tracked music, freestyle gfx, ....)? I haven't seen data like shown earlier by spike about these compos, but I doubt it's the livevoting that's borking, as Red Heaven was shown last.
If anything, the "preferred by pouët"-entries were shown relatively early in the compo (iks' piece was first I think, and The Spiral was somewhere around the middle iirc?), so it might just be the audience somehow forgetting them? I'm absolutely not sure of this either, though.
added on the 2019-08-22 13:56:20 by porocyon porocyon
or mayhaps the running order affecting how many stars you'll assign to relatively 'good' stuff early on in the compo as you're expecting the quality to improve. i myself have always been rather skeptical about the importance of the running order of things as 'quality will prevail' yet this evoke it seems to have mattered significantly
added on the 2019-08-22 14:17:27 by maali maali
Quote:
i myself have always been rather skeptical about the importance of the running order of things as 'quality will prevail' yet this evoke it seems to have mattered significantly

Yes, at Evoke 2017 the tune that was played first in the streaming compo actually won.

But with regards to the voting, participation at 62% means that pretty much out of three people you met at the party, one didn't vote.

Vote, please, vote. During the compo or after. Watch the compos and vote, otherwise one day you'll end up at a party with just a BBQ and livesets but no prods to watch.
added on the 2019-08-22 18:11:42 by LiSU^TRS LiSU^TRS
Maybe another reason for the strange voting results is that the guys who had collected >30 vote keys only cared to vote for the (crap-) releases of their own group? ;)
added on the 2019-08-23 11:24:46 by hfr hfr
@hfr: who did that? o_O

@LiSU: 62% is a pretty good score given that quite a bunch of visitors are just there for an afternoon or evening and leaving earlier, and not to forget non-scene friends which are also present and prefer not to vote.

But yeah, please vote!
added on the 2019-08-23 21:45:00 by T$ T$
Quote:
on behalf of the evoke organizing team,steam
Thanks for clearing that up and the insights. Much appreciated!
Already looking forward to next year.
added on the 2019-08-24 06:55:27 by numtek numtek
Pictures won't make it justice, but I tried - Here's a massive photo album for you! Thanks for being awesome as always :)

(As always, you have my full permission to use the pictures if you want to. Crediting is encouraged, of course.)
added on the 2019-08-25 14:43:16 by darya darya

login