basic steam economy

[Javert]

Originally posted by dscowboy
If Steam works it's going to be the Guttenburg Press of the gaming industry. [/B]

That's like saying that the Segway was the greatest invention since the internet. I might even put that in my sig as either prophetic or ironic.

Don't start giving it too much praise, those servers will come at a cost (banner ads+extra fees anyone?) and if we're to start comparing its server ability to that of fileplanet, you'll also be talking about fees, exclusives, and queue lines. THINK BUSINESS! Yes, part of Steam is for consumer satisfaction. But they're doing it for the future and for profit, not magnanimity. As of now and the next five (not ten) years, there WILL be bandwidth issues. Pre-caching still comes from d/l-ing from servers, a majority still in the US for an international community.

Yes, game devs will love sending out their work without BigMegaCorp Inc., but the BigMegaCorp Inc. will be Steam, since it, in a sense, published the game. And then more publishers (call them SteamII) will rise in competition. So in a sense, instead of BigMegaCorp Inc. Vs. BigMegaCorpII Inc., it'll be Steam Vs. Steam II, so game devs will once again luck out, percentages to Steam and Steam II.

The pros of Steam are anti-cheat and anti-piracy. The latter is what they're hoping for, since there will be a lack of hard-copies.

The concept is good, very good, but enough errors may just as well pull the whole thing down. And if it starts angering customers by limiting their choices in a market, it'll be Microsoft/RIAA/GameSpy all over.

 

[chadjones]

Originally posted by dscowboy
1) Publishers and retailers will crap their pants. They've just invested hundreds of millions into the rapidly growing game industry and suddenly they'll be stareing their own obsolescence in the face. They will start inventing heavily in their own content distribution networks.
2) Developers will drool at the opportunity to take 95% of the revenue from game sales, instead of the 10% they get now.
3) Independent developers will finally have the opportunity to publish without being sponsored by HyperMegaGlobalCorp Inc.
4) Game pricing options will become more flexible (like we've seen already with HL2), and we'll be able to buy smaller, indy games for a lot cheaper.
5) Competitors to Steam will pop up, like Yahoo Games on Demand. Except they'll be able to get 'first run' games on their network instead of the 2-3 year old stuff Yahoo publishes now.


If Steam works it's going to be the Guttenburg Press of the gaming industry.

I completely agree.

SubCog, how does reducing the cost of production result in "Inconveniencing or completely screwing the consumers"? Eliminating the middle-man distributer just means more money for Valve, which means Valve has more money to hire skilled employees/make more amazing games, etc...

I for one think Steam is good for the consumer and the developer.

 

[Javert]

Originally posted by chadjones
I completely agree.

SubCog, how does reducing the cost of production result in "Inconveniencing or completely screwing the consumers"? Eliminating the middle-man distributer just means more money for Valve, which means Valve has more money to hire skilled employees/make more amazing games, etc...

I for one think Steam is good for the consumer and the developer.

One possible way (of course is debatable) can be seen in my post right b4 yours.

 

[NothingMan]

Originally posted by SubCog
I offer a scenario: The first update for Half-Life 2 is released via Steam. Steam servers are immediately overloaded. Valve is unable to supply adaquate bandwidth for 3 weeks. Meanwhile, no one is able to authenticate Steam, because their not running the update. League matches cannot be held. Internet servers are empty. Lan parties will have to play Quake3.

I would also like to offer a scenario:

Little Joey Mc Johnson has just returned home from a long day of barely getting by in his remedial math and english classes at his school. His world view consists of a massive conspiracy to bring him down, and therefore nothing is his own fault. He seeks to lash back at the society that has scorned him so through cheating in his favorite new online game, Counter Strike 2.

However, this day, a small patch was applied virtually seemlessly to his Half-Life 2 executable. When he goes to load his HL2 Cheat-Loader, he gets an error message, and his cheats are not activated. Despite this, his need for release overpowers his need to cheat, and decides to play anyway. This newfound sense of using his own effort to achieve his own rewards reveals to himself that it is worth it to play by himself, and with his own skill. He casts away his cheating ways, and puts forth similar effort in everything he does. He starts to accel in school, gets a 1600 in his S.A.T.'s, attends an accredited yet resonably priced university, achieves doctorates in bio-informatics and genetics, and cures cancer and male patern baldness.

You really want to take away both little Joey's future, and cures for both cancer and baldness, all because you have to have a connection to the internet for two weeks to play Singleplayer Half-Life 2?
You're so selfish...

 

[Maskirovka]

Originally posted by ZeoviZ
Bandwith will not be an issue when bittorrent-like thing is included in steam. Until then, Valve has absolutely no chance of serving millions of users at the same time.

that's why they came up with their precaching idea...they send the game to you and unlock it later. this way they can avoid that initial rush of downloads (like when the non-beta client of steam got released)

so no, they don't have a chance of serving millions of users at the same time, but until they get the p2p stuff going, the precaching thing should work a lot better than the initial release of steam.

of course...the initial release of steam was a new client and that whole thing couldn't be avoided...but now that steam is established (not perfectly, mind you) things like precaching will be a good solution

EDIT: subcog, precaching also solves your "first update" scenario

 

[Wilson502]

how much bandwidth do u guys think valve needs to serve a hl 2 launch right now they have 1.5gbps of total bandwidth. If they are gonna launch hl 2 they are gonna need at least 3 gbps of bandwidth what do u guys think?

 

[Maskirovka]

http://www.halflife2.net/forums/showthread.php?s=&threadid=8159

they don't need that much

 

[Wilson502]

well think about it there is gonna be thousands of ppl buying hl 2, if not by the hundreds of thousands. Even if 1 or 2 out of 100 ppl get it through steam ur talkin at least 10 20k users downloading. You are gonna need a lot of bandwidth for something like that.

 

[dscowboy]

Originally posted by Javert
Don't start giving it too much praise, those servers will come at a cost (banner ads+extra fees anyone?) and if we're to start comparing its server ability to that of fileplanet, you'll also be talking about fees, exclusives, and queue lines. THINK BUSINESS! .

Arg... did you read any of my earlier posts? I'll repeat myself, it seems like the only way to get a point across on these forums is to post the same thing every page so people will actually read it.

Steam is going to be WAY overfunded. If you assume a cost of $500 per month for every 10 Mbps of bandwidth (that includes hardware and rack lease, and ISP fees), that means you're paying $500 for a max of 3.2 Terabytes per month (10 Mbps * 60 seconds * 60 minutes * 24 hours * 30 days / 8 bits per byte) If 200 people pay $50 dollars for a game, and Valve takes 5% of that to fund Steam (as they've said they would), that pays for all the bandwidth right there (200 * $50 * .05 = $500). 200 people * 2 Gigs = 400 gigs; only 1/8th of the bandwidth available for the month.

Selling 2 Gig games for $50, Valve will be collecting approx. EIGHT TIMES what they need to fund the servers. EIGHT TIMES. Fileplanet is trying to survive by getting PENNIES for each download via ads and charging a few people 7 bucks for unlimited downloads each month, Valve will be collecting so much money via Steam they won't be able to spend it fast enough.

These are the economics of Steam. I don't know where you got the idea that Steam was going to have to fund itself via 'queue lines'. People will be PAYING for the content they download, these are real games, not demos and trailers.

Even if Steam made no money at all, it would still be a huge profit center for Valve. Selling games on the shelf they get 10 - 15 bucks per full-price game. Selling them via Steam they get ALL the revenue. $50 - $15 = $35 per sale. So if 50,000 people buy via Steam instead of via retailers in the first quarter, Valve has just pocketed 50,000 * $35 = 1.7 MILLION more than they would have without Steam. It will cost NOWHERE NEAR 1.7 million to fund Steam for 3 months. The advantage of Steam is obvious, Valve can't afford NOT to invest heavily in it.

 

[Maskirovka]

Originally posted by Wilson502
well think about it there is gonna be thousands of ppl buying hl 2, if not by the hundreds of thousands. Even if 1 or 2 out of 100 ppl get it through steam ur talkin at least 10 20k users downloading. You are gonna need a lot of bandwidth for something like that.

did you read the thread i posted the link to?

they're going to spread out the downloads over a week or so...so everyone that preorders it will actually have it on their computer (encrypted) and it will be "unlocked" on the release day. this avoids that whole inital rush thing.

it won't be perfect, and i'm sure the servers will still be strained, but nowhere near as much.

 

[dscowboy]

Originally posted by Wilson502
well think about it there is gonna be thousands of ppl buying hl 2, if not by the hundreds of thousands. Even if 1 or 2 out of 100 ppl get it through steam ur talkin at least 10 20k users downloading. You are gonna need a lot of bandwidth for something like that.

Well do the math. Let's say 10,000 people 'preorder' it through Steam and Steam starts downloading it to them a week before release. 10,000 * 2 GBytes = 20 TeraBytes. With 1500 Mbps, they have a total capacity of 1500 Mbps * 60 sec/minute * 60 minutes/hour * 24 hours/day * 7 days/week = 907 terabits in a week. 907 Terabits / 8 bits per byte = 113 TeraBytes, 5 times what they need.

Of course, there's always overhead, that eats up a percentage of bandwidth, and there's the fact that their servers won't be at max usage all week, only during peak hours when everyone is preloading at once. And nobody really has any idea how many people will buy via Steam and pre-load it. So I agree, I think they should increase the amount of bandwidth available, get the other servers online and double it perhaps.

 

[Rubeus]

I do believe we have started an intelligent discussion on this forum. I was starting to think it was impossible! DSC, you make good points. Threadstarterguy has good arguments, but each point definitely has counterpoints. I am currently very pleased with Steam, I absolutely love the interface and the friends list. I do however understand the inconveniences it could bring to some gamers.

Let me speak theoretically here for a moment, since I have no numbers handy...

I hear the argument that 56k users are going to have a bitch of a time with Steam. I have played fps's multiplayer on dialup before, and I must say, what's the point? A 56k user who wants to play multiplayer will have to deal with Steam. A 56k user who wants to play singleplayer only will have no problem.

And my ignorant addition to this argument for the night:

You can't expect to play HL2 on a 486, right? Technology has advanced by leaps and bounds past those old relics. Can not the same situation be applied to 56K / Broadband connections? This may sound cruel.. but if you expect to be able to play the newest, most badass 3D game online, then get with the damn times or suck it up

G'nite!

-Rubeus

 

[raloth556]

I like steam

I have really fast internet and love to use it. I will preorder through steam and hopefully through the preaching scheme thier using, have it by the release day. I get close to 3mb/sec downstreams, I download full games and demo’s regularly and it takes no more than a hour or 2. Valves server bandwidth has had it's ups and downs but there have been times where I was able to download the half life vids at really good speeds. For people that already make large downloads on a regular basis steam is a great utility, it's better than searching the internet and waiting in line to download. I want steam to automatically download everything Half life related while I’m not playing or doing something else on my pc. That way when I’m ready to play I will have everything already as apposed to having to wait to play. Not to mention the fact that I hate having to have the cd in order to play and I have several different computers, steam will allow me to play on all these different machines, some of which are at different locations without the need of a cd.

I agree that steam can not replace cds for many many years due some of the limiting factors you stated. All the normal media delivery mechanisms are still in place for half life 2, in my opinion people who do not like steam or do not meet the requirements for successful utilization of steam should buy the cd and not think twice about it.

 

[SubCog]

good arguments, everyone. You are right about many issues here.

However, I still have problems with the idea of being forced to connect to the internet to play single player and lan games, even if for only a couple weeks. It makes me wonder "why?". There doesn't seem to be any reason behind it.

Except of course if you buy it via steam in the first place. then it makes total sense, since you don't have a cd key and you obviously have adaquate internet.

But I also take issue with not being able to manually download updates. If you play single player or Lan on a closed network, you can't bring in a patch on a cd. I cringe when I think of Lan parties where not everyone has got the newest patch yet, and there's no way to sinc the game.

And as for bandwidth, I do believe that eventually valve will offer enough. However, I think that those of you who think most people will be able to cache everything in a week are living in a valve pipe dream. Right now anyway. But, yes, this will be smoothed out over time.

Really, I know that steam features are the wave of the future. I just think that standardizing everything with steam right now is a bit pre-mature. I wish they'd ease it in slowly... allowing the community to participate voluntarily... and over time phase out the other stuff.

 

[Wilco]

My 2 cents.

For the "cant use fast PC to download updates 2 transfer to 56k PC". What about Cache files? We've already seen this with the Relaesed Steam versions with the games already included. This can easily be pass around on a CD or floppy if its small enough. This also solves the LAN thing.

As for options about LAN Authing. I haven't seen anyone mention the idea of local Auth servers for LANs. If when Hosting a LAN game it also sets up a local Auth server that checks peoples SteamIDs to make sure no 2 people are playing on the same LAN. This wouldnt rule out copying but will hinder it alot, people generally copy games for friends who they then play against. Making the SP Game need CD Auth only will be okay. And obviously if its been downloaded from Steam they have a good enough connection to allow internet connection checking at every start.
The couple of weeks would give them time to make this.

Its got 2 happen sometime, why not now? They could wait until HL3, but some-one else might have done it by then.

Pre-caching and constant updating should solve the "Patch server overload" problem.

If your going to pre-cache your gonig have a Always on Internet connection. Either leave the PC on all the time or turn it off only at night and it'll be fine.

to download 2Gigs (2,000,000 Kbytes ?) at 5kbps (thats a 56K speed) will take:
400,000 seconds
6,667 minutes
111 hours
4.63 days

(I hope I converted 2Gigs right )

assuming 1.5Gbps bandwidth they can serve 300,000 users at that speed.

 

[Wilco]

Oh and the voluntary community involvement part was 1.6 and Steam Beta

 

[|\/|oDdiC(_)$]

ouch, and good research

 

[poseyjmac]

Originally posted by Rubeus


You can't expect to play HL2 on a 486, right? Technology has advanced by leaps and bounds past those old relics. Can not the same situation be applied to 56K / Broadband connections? This may sound cruel.. but if you expect to be able to play the newest, most badass 3D game online, then get with the damn times or suck it up

G'nite!

-Rubeus

but see, i know people that WANT to get broadband, and HAVE enough money. the problem is they are too far away from DSL range. also no cable modems in our smaller city. so this leaves them with either satellite or isdn. satellite has slow ping times. isdn has better ping times then dialup, but costs WAY too much, plus the per minute BS charges.

the only way for some people to get a good connection is if they move, and thats asking too much from normal people.

 

[SubCog]

actually, ethernet protocols really only transfer at about 1/10th of the total bandwidth. It's the way ethernet sends packets. Who here has a 1000mbps connection? When was the last time you got a 1000mbps download. It doesn't happen. You'll be lucky to get 200kbps.

So I think its safe to take all the math and multiply it by ten. 4.6 days turns into 64 days.

 

[Wilco]

wtf? I started with 5 kbps!
I can get 50kbps off my Broadband connection Easy. So skip your multiply by 10 part, we're back to 4.6 days

 

[Wilco]

Thats using real world data, not theoretical maximums Tho you could say it can only service 30,000 users

 

[Baggins]

Mentioned here earlier was someone complaining about the lack of ability to patch the game via CD's or something when at a LAN.

To you I just want to say one word: THINK!

For all you others, here's how that'll work: Everyone starts up thier game, and takes notice of what version he/she is running. THEN, the person with the highest version number either copies his/hers cache files to CD, or shares the folder with the cache files on the network (you'll be at a LAN, for chrissake!), then everyone else copies those files, and suddenly the entire room is synced with the latest version of HL2....

For those with the CD/DVD version, there is ALWAYS someone that has dl'ed the latest patch before the LAN. OR, be smart and do it yourself...

I think that should be easy enough to understand...

 

[Guest]

I have just one question about the pre-cahcing idea.

That really only works assuming people have steam and their internet connection running full time and with significant downstream concurrently.



If you are playing a game online, are you really going to want a download stream going that is "precaching" your game and raising your ping?

Trade off for "faster" patches, I suppose.

But to say that people are going to magically turn on their computers and then notice they already have the patch that came out today because of precaching is stretching things quite a bit.

 

[Baggins]

Originally posted by Sh0t
I have just one question about the pre-cahcing idea.

That really only works assuming people have steam and theri internet connection running full time and with significant downstream concurrently.

If you are playing a game online, are you really going to want a download stream going that is "precaching" your game and raising your ping?

Trade off for "faster" patches, I suppose.

But to say that people are going to magically turn on their computers and then notice they alreadyhave the patch that came out today because of precaching is stretching things quite a bit.

That feature can be easily turned on and off in the Steam settings. If you're going to play online, just disable the automatic downloads, and re-enable them when you're done.

Maybe it's just because I'm not anal-retentive, but I don't see any problems with Steam at all as long as the servers are up... Just turn on your computer in the morning, and let it run downloads while you're at school, uni, work or whatever...

Ease up, people!

 

[Guest]

I don't have a problem with the idea of Steam, but I think it's only fair to comment on both sides of it. A lot of the pro arguments involved precaching, which is a good idea.

But it's not without a price.