what you could do is upload the file once to a fast server
and then remote up it to different 1click hosters
that way you can move gigs of data in some minutes
http://www.1-click-hoster.dl.am/ here is a good overview of hosters with remoteup
(its german but not so difficult to understand i think)
to pick one out: http://www.fullshare.de/ always seemed to be fast and stable for me (upload limit 1gb)
quick edit: of course there are lots more 1clicks around who support remote uploading
if you plan to make it that way i would have no problems to search some more for you
I think xvid would look a lot better considering the quality you get with movies that are 1.5hours long and ~700mb, and 42minute tv episodes are 350mb. 500mb you could fit 40minutes in with excellent quality. The only downside would be the encoding time as which could take a while to compress with 2 pass, I'm not sure how long wmv takes as I've never used it.
Another way to distribute the files would be bittorrent not sure if you've looked into it but using that could save on the hosting bill if you're upload is good enough and other people share the movie also which I'm sure they wouldn't mind sharing their bandwidth. This could be used alongside the http download.
Currently bit torrent is fine, but ultimately it makes for uncatalogued downloads with unprovable stats and would not be useable to attract sponsorship.
I can't see how that's different to the youtube or google streaming? Most if not all trackers provide a number of how many times the file has been downloaded.
@Davo: Have a look at the Google Video and You Tube versions and they both show the number of views, it's provable to a sponsor. I am not much of a p2p user, but my understanding of it suggests bit torrent does not track downloads. I'm totally confused by your statement?
800x600x16. It gets converted 3 times before being uploaded sadly, but followed the same process as the other 2.
There is a 'snatched' statistic on the tracker that counts how many times the file has been download 100%. This number represents 100% downloads not just hits and disconnects so it'd also be a good way of keeping track how many people download via bittorrent. That's how you could tell how many times it was downloaded. It's the same as youtube etc just showing a number of downloads or views.
Torrents have a large advantage in getting a file out to many people quickly. If 10 people start downloading from the same server then it could slow things down, but with torrents everyone is sharing the file so it's a lot quicker the more people that jump on. If everyone wants the video straight away it's a great method. There's even a way to get a server with a high upload speed ont he torrent to distribute it even quicker.
great video, commentary was the best so far (sorry Mike and Tristan , I hope you are going to even better this with round 4 ) due to the thing someone mentioned before that Becky managed to show some real enthusiasm for the actions on the track, was very lively.
The radio sounds were funny when I first heared them, but they are a good addition as were the sounds from the start/finish straight camera angle.
Becky I was wondering if you used any AA and AF when recording, because the edges seemed too ragged in the video, also a try with XviD would be worthwile, it should be possible to improve the quality of the video further.
looking forward to the next release as I will probably feature in it a little more than in this one
what is 1337? and what does leet mean at all? and what is a fair bit? I consider below 4xAA and 16xAF is still to ragged. if Becky's video card can't take more then let's get a good cause up to get a decent GFX card for her
I use twin overclocked 7800GT 256mb's, but a lot of people commented on raged edges this round, i'm wondering if I didn't have the correct AA/AF settings when I recorded it, perhaps because I had been playing another game.
My CPU is a little week, actually it's pretty good but what i'm trying to do with it (full screen Fraps in realtime) means I cannot quite get the resolution I want. The output is 740x576 NTSC at 30fps, so I record at 800x600, but I cannot get 30fps in 32bit, so I have to use 16 bit. My motherboard can take a better CPU - I actually alwaysplanned a CPU upgrade before XMAS when I bought the machine earlier this year - but I cannot afford it (which wasnt part of my plan).
I'm not sure if a little extra CPU juice would be enough to get 32bit encoding, anyway the processor is an AMD 4400 dual core. It's not bad, but I think each core works slower than my old 2Ghz chip. The key here is it's on an 883mhz bus using dual channel ram, because when capturing a lot of data is going through the bus - I cannot make that any faster without liquid cooling.
EDIT: The other thing which effects Fraps performance of course is the speed of the hard disk. I use a striped RAID array with 2x SATA-2 hard disks, theoretically 6gb/sec transfer rate. However I bought Hitachi drives and have since learned from experience that Hitachi are not fast. Overall I would say the performance is better than having a single drive, but RAID-0 should theoretically be double the speed (there's no overhead when using identical drives).
well your system should be more than enough to get decent quality videos.
there are quite a lot video makers in our hungarian forums and the experienced ones suggested recording at lower speeds (0.25 gives the best frames, but 0.5 would also do I guess) to avoid the real time fullscreen fps problems. Also you should succeed with 32 bit this way as well. worth a try I guess
when doing the editing you just need to speed up the playback to get real time speed. this will result in a lot smoother playback, also with the new patch slower playback will also produce good sounds, so there is really no drawback here, except for the more time consumed capturing the scene.
yes, I thought it was a lot of time, but you could just do a comparison with a 30 sec or 1 minute clip to see if there is any significant difference or not and consider later if it would be worth the extra time at all.
*cough* lower multi - raise FSB, but memory on a divider - aka Overclock! *cough*
My 4000 is running at FX-57 speeds, and is running 17% faster than stock using air, with the stock cooling you could get 10% overclock easy!
well, i run 4xAA, and when i turn it up to 16xAA i can't see a difference while racing, and it really hurts my FPS, so i don't!
well you must have one hell of a GFX card if you can run 16xAA
AA is anti aliasing, AF is anisotropic filtering, with my x1900 gt I can set max 6x AA and max 16x AF.
if you mean compared to 4xAF 16xAF doesn't bring much of a difference but fps drop I can understand that (although I am not sure what does AA and what does AF with the graphics), try to use AA as well to see the difference visually. If you can'T see any difference between no AA and no AF and 4xAA and 16x AF something must be wrong with your settings. maybe they don'T work at all, I had that before and it turned out I was using the wrong tool.
I know what AA does, and i said i can't see the difference between 4 and 16 while racing (although in screenies it is pretty obvious!), but yes i can tell the difference between 0 and 4x!
Thank you for the suggestion, I never thought to look into this.
, I hope you are going to even better this with round 4 
