PSN game bloat?

Absolutely, I wasn't arguing against compression. I was merely indicating that if it's being used at all, it's not being used very well.
They better be using it, I'd think 1 MB over 78MB is significant in terms of BW usage and acceptable for generalized compression algorithms working on already compressed data.
That's an easy problem to solve, though. The hypervisor for each box gets a key that is used to sign a portion of the downloaded binary after it has been downloaded and validated. Until the box is signed with the local key, it's basically considered an NX portion of disk. It's a simple three-way handshake.
EDIT: Actually, based on your post, it sounds like this is exactly what they do already.
yeah, that's what I expect them to do.

I can say it. No consumer would stand for it.
I doubt it will be a problem for a free service (demos, trailers and home stuff), especially with a suitable EULA section.
The only reasonable option is to make this an opt-in proposition, and you would never get enough people to opt-in to get any value out of something like this.
For priced downloads, you can motivate users with free or cheaper content.
I don't think it's a problem. My problem is that the user is required to take an explicit step to run an installer.
I know, my question was merely a transition from difficulties in streamed installation to two stage transparent installation.
Fortunately, modern architectures allow for prioritizing both IO and CPU usage. There's no reason why the install stage can't be put on low priority so that they'd simply get starved in the case that a game needs to do work.
We are not talking about a general purpose FS and CPU usage, we are talking about a specialized gaming machine and its needs.
If I were them, I'd definitely optimize FS for single continuous writing especially with DRM.
Hell it may even be due to memory constraints of OS.

In any case, whether those contribute to a factor or not, they could always provide autoinstallation outside of games or even, which definitely tells something about design.
 
They better be using it, I'd think 1 MB over 78MB is significant in terms of BW usage and acceptable for generalized compression algorithms working on already compressed data.

Very true. 1MB is cheap, 1 MB over 200,000 downloads suddenly becomes much more interesting.

I doubt it will be a problem for a free service (demos, trailers and home stuff), especially with a suitable EULA section.

For priced downloads, you can motivate users with free or cheaper content.

Sticking something that many could consider a privacy violation (not to mention people who might get charged for the up without knowing it) in a EULA is a great way to make people angry. It's been proven time and time again that people don't read EULA's, and even if they did, they're generally written such that an average person couldn't understand it. They don't hold up in court, and they certainly don't hold up in the press. I kind of hate the concept of EULA's in the first place. They're trying to put a legal burden on par with buying a car on something more akin to buying a potato. It's not reasonable, and the courts (at least in many countries) agree.

I think the idea of providing a financial incentive to allow your box to be a node is interesting. I'd be curious to know how much money Sony could save and how much of a discount they could actually provide. How to explain what it means and make people aware of it is another problem, but not an insurmountable one.

I actually think this is a really interesting idea for a lot of reasons.

In any case, whether those contribute to a factor or not, they could always provide autoinstallation outside of games or even, which definitely tells something about design.

Indeed. It might turn out that by prioritizing IO/CPU, it would be the case that the install would be completely stalled during gameplay. That said, the 360 has to deal with the same problem, and they don't seem to have any problems on that front. My guess is that the processing costs are limited, and the sliver of IO needed to sign the binary just doesn't rate in the avalanche of data that a game is dealing with most of the time.
 
I can say it. No consumer would stand for it. The only reasonable option is to make this an opt-in proposition, and you would never get enough people to opt-in to get any value out of something like this. Even then, it's a political landmine. Unless someone gives Sony some really bad advice, they wouldn't touch this sort of thing with a 10-foot pole.
Konami is already using torrent for the distribution of MGO beta along with HTTP. Regardless of political issues, they have to offer an HTTP option since ISP may filter torrent or throttle upload.
Sticking something that many could consider a privacy violation (not to mention people who might get charged for the up without knowing it) in a EULA is a great way to make people angry.
That's an interesting take, however multiplayer games upload data too and I've not heard of such a EULA item about bandwidth cost. As for privacy, sure you can know IP addresses of other peers who are downloading the same file, but is an IP address privacy that is tied to someone's identity? It's just a number that belongs to an ISP. Oh and you can know IP addresses of others who are playing the same game in Xbox Live too!
 
Konami is already using torrent for the distribution of MGO beta along with HTTP. Regardless of political issues, they have to offer an HTTP option since ISP may filter torrent or throttle upload.

Could you give a few more details? Do you have to explicitly request to do torrent for it? Also, what's the use been like?

That's an interesting take, however multiplayer games upload data too and I've not heard of such a EULA item about bandwidth cost. As for privacy, sure you can know IP addresses of other peers who are downloading the same file, but is an IP address privacy that is tied to someone's identity? It's just a number that belongs to an ISP. Oh and you can know IP addresses of others who are playing the same game in Xbox Live too!

A multiplayer game is pretty explicit though. It's a peer-to-peer, or at least server-to-multiple-peers system by default. It's not really a good comparison here.

When you are simply clicking something and downloading from somewhere, I think it's fair to say that there is a reasonable expectation that you will NOT be connecting via peer to other users by default. Or worse, simply having ever downloaded a game and connecting to PSN turning you into a node. These are the sorts of things that are potential problems.

As far as IP addresses being privacy data... it's a sticky point and how private it is really depends on application.
 
Could you give a few more details? Do you have to explicitly request to do torrent for it? Also, what's the use been like?
Well you can check out the thread... users can explicitly choose P2P or HTTP. It can be faster to download than HTTP for some users as long as peers are all in the same country with good network infrastructure.
A multiplayer game is pretty explicit though. It's a peer-to-peer, or at least server-to-multiple-peers system by default. It's not really a good comparison here.
Privacy-wise, it's comparable or more significant. In fact, it's possible to associate a user ID with an IP address in a P2P game connection.
 
Well you can check out the thread... users can explicitly choose P2P or HTTP. It can be faster to download than HTTP for some users as long as peers are all in the same country with good network infrastructure.

As long as you have to make an explicit choice to do it, there's no problem on that front.

Privacy-wise, it's comparable or more significant. In fact, it's possible to associate a user ID with an IP address in a P2P game connection.

Yes, you certainly are exposing more private data while playing a multiplayer game, however, this is a necessity of multiplayer and is likely to be a surprise to no one. In the case that simply downloading a file exposes data about you, the reality is no longer in line with reasonable expectation.
 
Back
Top