Meters for downloads of speedruns

2006-05-19 11:01:54 am

piece of cake!

Hey everybody, I don't know if this Idea is possible or already asked on this topic (don't have time to read all...)
But I mean that we can put a meter like Internet Archives, to know how many times a speedrun has been downloaded, and what the average, the rating...!

that would be interesting I think! Grin

Thread title:

Zhegan

2006-05-19 04:22:18 pm

[Insert text here]

If it's possible it sounds like a really good idea, shouldn't be to hard, but then again I don't know that much about how the internet works

darkwasabi

2006-05-19 05:15:34 pm

King of hearts

Why bother, as you've said, Internet Archive has one. It just means more work for no reason.

spineshark

2006-05-19 05:24:49 pm

新世紀進歩的羽扇子音楽

archive counts wrong though. still seems kind of pointless to me.

tagh_azog

Edit history:

tagh_azog: 2006-05-19 05:28:29 pm

2006-05-19 05:25:12 pm

piece of cake!

yes, but speeddemosarchive haven't all their speedruns which can be downloaded on internet archive, so all the downloads are not from Internet archive... :-/

Also, I would like to know how many people are interested by speedruns and how many people download my speedrun for example!
why do you find it pointless? I just ask, if the work to do it is not so boring...why not?

darkwasabi

2006-05-19 05:37:18 pm

King of hearts

Quote:

why not?

The same reason why 99% of simple suggestions don't get accepted: more work.

Internet Archive's counter, I think, works by counting how many complete files are transfered. This includes High and Insane Quality, which take up 90% of the whole thing, and not many people download these.

But having this isn't really beneficial at all, especially compared to some other suggestions that are.

Zhegan

2006-05-19 05:52:25 pm

[Insert text here]

Quote:

The same reason why 99% of simple suggestions don't get accepted: more work.

Internet Archive's counter, I think, works by counting how many complete files are transfered. This includes High and Insane Quality, which take up 90% of the whole thing, and not many people download these.

But having this isn't really beneficial at all, especially compared to some other suggestions that are.

Does archive.org count 1 download per segment for segmented runs? because it look like they have a ton of downloads on speedruns like that :S

darkwasabi

2006-05-19 06:14:45 pm

King of hearts

Quote:

Does archive.org count 1 download per segment for segmented runs? because it look like they have a ton of downloads on speedruns like that :S

Can't be, look at MGS3 for the segmented and SS. The SS has slightly more downloads because it was here first. If it counted each file downloaded, the segmented run should have 10 times more downloads.

AjAX

2006-05-19 08:02:12 pm

Quote:

The same reason why 99% of simple suggestions don't get accepted: more work.

Internet Archive's counter, I think, works by counting how many complete files are transfered. This includes High and Insane Quality, which take up 90% of the whole thing, and not many people download these.

But having this isn't really beneficial at all, especially compared to some other suggestions that are.

Why making a SUGGESTION folder in that case if 99% are not accepted youd rather closed that up.

darkwasabi

2006-05-19 08:13:56 pm

King of hearts

Quote:

Why making a SUGGESTION folder in that case if 99% are not accepted youd rather closed that up.

I dunno, ask Radix?

But anyway, what I meant by that was that most suggestions aren't worth the work needed to implement them. They're just minor details that only few would actually use or even notice. This is a useful suggestion, but there already is a system set up by archive.org that's very good. The only other place that may "steal" downloads is dream hosters, but it's not always used. And you can't count how many downloads there are from torrent files.

nate

2006-05-19 08:18:38 pm

sometimes we get good suggestions, i dunno. as for the download counter, it would be almost trivial for radix to put that in, but it's not something he wants to do, for two reasons (radix, correct me if i've got any of this wrong/skewed).

the primary one is that we can't actually measure downloads. runs are often released in multiple formats on multiple hosts using multiple services: h264 iq vs hq vs mq vs lq vs divx mq vs lq, tph server (speeddemosarchive.com) vs sda.dreamhosters.com vs archive.org american and european servers, and http vs ftp vs torrent. add on to that the use of download accelerators permanently damaging download counts we do have access to and you see that we really have only a vague idea of what people are downloading.

the secondary one is that even if we could count downloads, we wouldn't let people know how popular runs are, because we don't want people having popularity contests with their runs. the run is the content, not the runner. if a run is very popular (downloaded a lot of times), the run will not change or become annoying on the forum, while the runner will if he or she is told that they are popular/essential to sda. this is not a situation we want to have on our hands, especially when it's one we directly caused. every run and every runner is essential to sda. no one run is more important than another, at least not in my eyes.

sometimes we get good suggestions, i dunno. as for the download counter, it would be almost trivial for radix to put that in, but it's not something he wants to do, for two reasons (radix, correct me if i've got any of this wrong/skewed).

the primary one is that we can't actually measure downloads. runs are often released in multiple formats on multiple hosts using multiple services: h264 iq vs hq vs mq vs lq vs divx mq vs lq, tph server (speeddemosarchive.com) vs sda.dreamhosters.com vs archive.org american and european servers, and http vs ftp vs torrent. add on to that the use of download accelerators permanently damaging download counts we [i]do[/i] have access to and you see that we really have only a vague idea of what people are downloading.

the secondary one is that even if we could count downloads, we wouldn't let people know how popular runs are, because we don't want people having popularity contests with their runs. the run is the content, not the runner. if a run is very popular (downloaded a lot of times), the run will not change or become annoying on the forum, while the runner will if he or she is told that they are popular/essential to sda. this is not a situation we want to have on our hands, especially when it's one we directly caused. every run and every runner is essential to sda. no one run is more important than another, at least not in my eyes.

Zhegan

Edit history:

Zhegan: 2006-05-19 09:27:55 pm

2006-05-19 09:27:23 pm

[Insert text here]

Quote:

Can't be, look at MGS3 for the segmented and SS. The SS has slightly more downloads because it was here first. If it counted each file downloaded, the segmented run should have 10 times more downloads.

I don't want to sound like a moron, but how DOES archive count segmented runs anyways? does it divide by the number of segments? because I'm sure there have been times that someone downloaded a few segments then decided not to download the rest.

ballofsnow

2006-05-19 09:32:20 pm

Could be done by unique ip address.

darkwasabi

2006-05-19 09:57:23 pm

King of hearts

Like I said, it probably counts how much was transfered.

Ekarderif

2006-05-19 11:12:28 pm

Your almighty lordship

I'm not sure about storing IP addresses. It seems small, but SDA is experiencing explosive growth. Both the number of runs and the number of users are increasing exponentially

1 user downloads 1 run: 1 entry.
100 users downloading 100 runs: 10,000 entries.
(estimated current usage)
500 users downloading 245 runs: 122,000 entries.

I'm sure you can see where this leads.

nate

2006-05-19 11:21:37 pm

yeah, you know you've got a popular site when running out of disk space due to the web log getting too large is a constant concern.

Radix

2006-05-19 11:41:05 pm

I'm addicted to games

You shouldn't read into archive's download numbers at all, because I don't know exactly how they work even for items with one file, so clearly they're useless.

And as nate said, we have no intention of turning things into a popularity contest. Although I will say that Metroid runs are _probably_ more important than say, Chameleon Twist. Tongue

Enhasa

Edit history:

Enhasa: 2006-05-20 09:07:40 am

2006-05-20 01:57:41 am

everybody wanna tell you the meaning of music

This is something I agree 100% with Management. Wink

The run should speak for itself.

My biggest problem is that download meters would mostly reflect how popular the game is and not how good the run is. The very worst idea would be average rating like the OP said.

I don't know why people always seem to think archive's stats are worth going by, but there is or was some sort of "behind the scenes" unofficial counts. From what I know and think is worth/prudent saying, you mainly console gamers or forum regulars might be surprised (maybe not) that most of the top downloaded runs are PC games. I say maybe not, because it makes sense that PC game runs get linked from other specific game sites/boards, and many of these people post elsewhere and not here.

Quote:

if a run is very popular (downloaded a lot of times), the run will not change or become annoying on the forum, while the runner will if he or she is told that they are popular/essential to sda

Doesn't this happen right now though anyway?

Edit for nate (below): Well this was a tongue in cheek comment of course, but you are correct.

This is something I agree 100% with Management. ;) The run should speak for itself.

My biggest problem is that download meters would mostly reflect how popular the game is and not how good the run is. The very worst idea would be average rating like the OP said.

I don't know why people always seem to think archive's stats are worth going by, but there is or was some sort of "behind the scenes" unofficial counts. From what I know and think is worth/prudent saying, you mainly console gamers or forum regulars might be surprised (maybe not) that most of the top downloaded runs are PC games. I say maybe not, because it makes sense that PC game runs get linked from other specific game sites/boards, and many of these people post elsewhere and not here.

[quote]if a run is very popular (downloaded a lot of times), the run will not change or become annoying on the forum, while the runner will if he or she is told that they are popular/essential to sda[/quote]
Doesn't this happen right now though anyway?

Edit for nate (below): Well this was a tongue in cheek comment of course, but you are correct.

Zhegan

Edit history:

Zhegan: 2006-05-20 02:29:50 am

2006-05-20 02:26:32 am

[Insert text here]

Yeah, now that I think about it...people would probably judge how good a run is by it's download count, which would really screw over newer runs added, and any runs on games that arn't hugely popular...basically encouraging people to only download runs of Zelda, Mario, Metriod, and Quake (and of course a couple other franchises).

On the other hand ratings would be even worse, since one person would probably rate a run as 1/10 for no reason (yet they we all know they still would) causing everyone to ignore the run completely thinking it sucks without giving it a second look.

nate

2006-05-20 06:57:31 am

Quote:

Doesn't this happen right now though anyway?

all things are relative.