Archive for March 2009

Predictions for the Australian F1 Grand Prix (Part 2)

More notes. I’m already starting to doubt some of the details in the previous post. McLaren, for example, are being very downbeat about their chances. Apparently the car still has some issues.

Red Bull
I have a lot of respect for Adrian Newey as a designer, so I hope that the “beautiful” car that he has designed will also be fast. Webber’s injuries during the winter will have interfered with his preparation and I think that Sebastian Vettel will have a better handle on the car for the first few races. I expect RB to be ahead of Toro Rosso, but still mid pack.

Toro Rosso
I don’t really know very much about the STR car but I’ll include the team for completeness. I guess it is either last years Red Bull with updated aero, or perhaps they have managed to get hold of a few 09 chassis from Red Bull. I have followed Sebastien Bourdais since Champ Car and am glad he kept the seat at Toro Rosso. I don’t know too much about Sebastien Buemi either, I’m sure he has the potential to be quick, but as a rookie, he’ll probably need some time to settle in.

Toyota
I just read some news that Toyota have been able to keep their controversial “double” diffuser. This should give them a big advantage this weekend. Trulli and Glock are solid drivers, so they may be up around 3rd or 4th this weekend.

Renault
When I was over in Brazil last year, many of my friends told me that they didn’t like Nelson Piquet Jr (Nelsinho) because they felt he had never worked very hard for his success and instead inherited it from his father. I think Fernando Alonso is a great driver and I expect him to make the best of a mid-pack car.

I’m not really in any position to have any deeper insight into the performance of the cars, so all of this is just conjecture based on past performance, off-season news and other nebulous factors. Anyway, it was fun to write and it should be fun to see how wrong I was on Sunday.

OnLive

There has been some stuff in the news lately about OnLive. It is promoted as a sort of “thin console”, where all the game rendering, physics and other computation is done on a remote server and the video is streamed to a small box connected to your TV. The advantage is that the OnLive machine can be a very cheap piece of “dumb” hardware, requiring a low initial investment and very little maintainance or upgrades. I’ve seen the concept around several times before and I don’t really know how they plan to address the obvious challenges to this sort of model.

Firstly, bandwidth. Streaming unique 1080p video to a customer requires a lot of bandwidth, even transmitting real-time-compressed video would require several Mbps at least. Current consumer internet connections just can’t handle that kind of sustained bandwidth. I don’t even think that there are server-side solutions to the problem of sending 10s or 100s of thousands of simultaneous 1080p video streams, the server farm would be immense.

Which leads me to another point. There still has to be a machine capable of playing the game somewhere. If it is not in the players house, it must be at the server side. If the game features 1080p 3D graphics and realistic physics and the other hallmarks of next-gen games, then that machine is going to have to be pretty beefy. Even then, one Core 2 server with a modern graphics card will probably only serve one or (possibly) two clients. Serving games isn’t like serving websites, one server can’t handle hundreds of gaming clients. If 10,000 OnLive customers want to play some modern game like (say) Killzone 2, then there is going to have to be close to 10,000 PCs on the server side, creating all that content. Again, this sort of investment is not practical.

Finally, latency is just a killer in this situation. Anytime the player provides the system with any input, it must be sent up to the server, processed, and then the result sent back to the client as part of the video stream. This can make simple things like mouse interaction with menus an awful user experience. It becomes tempting to try and do some of these “simple” things client side, put a small processor in the OnLive box so it can handle menu navigation etc. That only really solves the simple problems, anything more complex, like client-side prediction in a FPS is impossible. Client-side prediction is the technology that makes FPSs playable over a network.

I think, in general, the remote-processing model is sound. It is just that the technical issues really limit the sorts of experiences that can be produced. I think the graphics have to be conducive to compression, ideally even vector based, the server-side computation must be able to be run as 10 or more instances on one server, and the game types have to be non-realtime to avoid the latency issues.

Predictions for the Australian F1 Grand Prix

I thought I’d just write a few lines about how I think each team will go in the Formula 1 season opening round this weekend in Melbourne.

McLaren
Have been accused of ‘sandbagging’ in recent testing sessions by some pundits, while others claim the relatively poor performance of the car is due to significant aerodynamic issues. They have confirmed the Melbourne car will feature a KERS. My feeling is that McLaren is a professional team and they should be on top again.

Ferrari
Nothing special to report here. They will be using a KERS. I expect them to be quick, Massa quicker than Raikonnen. I do like Raikonnen though, so it would be good to see him come back and have a great year.

BMW
Have suggested that they will leave the decision to run a KERS until as late as Saturday. Team boss Mario Theissen has suggested that the power boost advantage is offset by the extra weight and the impact on tyre wear. I don’t know why they would leave the decision so late, BMW is a high-tech team and must have modeled the effect of a KERS very accurately. Maybe they are still working bugs out of the system. I think that BMW is fundamentally stronger than Renault, but there is an outside chance they will be behind Renault this weekend.

Brawn GP
The former Honda team will have had a very disrupted off-season with the struggle to find a buyer. I have to believe that this will have slowed the pace of development of the car. Running Mercedes engines this year is a step up from the Honda engines of last year but the chassis/aero can’t have had enough development. I don’t know what they have been doing to have such great times in testing recently, but I can’t see it happening in the race. I’d love to be wrong on this one too.

Williams
Have already spoken about trying to do enough this year to retain the services of driver Nico Rosberg. This seems like a defeatist attitude. Sometimes large-scale rule changes such as those seen this year can provide an opportunity for mid-tier teams to make a step up, but I think that Williams is likely to stay about the same.

Force India
Once again bringing up the rear. They have had limited testing during the winter and driver Fisichella has already spoken of looking forward to a new aero package and KERS that won’t be available for the first few rounds of the season. Mercedes engines this year.

I’ll try to add some thoughts about the other 5 teams soon.

Mood Music

Two drum and bass tracks, the first takes me to a dark place and the second brings me right back.

What happened to 1UP?

Early this year, 1UP.com was purchased by UGO, and many of the gaming news podcasts created by 1UP were shut down. Sure, some of the great regulars from these shows had left over the years, but the opinions and discussions were still entertaining.

These days, many of the groups that had worked on the old 1UP podcasts have started up elsewhere. Check them out.

Some of the 1UP Yours guys have continued at 1UP.com, starting a new show called ListenUP.

The 1UP Show video podcast has been reborn outside 1UP.com as CO-OP on the Revision3 network.

There is also a new podcast called RebelFM, that stars many former staff of 1UP.com, though I must say I haven’t got into this as much as the other two.

Star Trek trailer

They did a really good job with the new Star Trek trailer. I got goose-bumps watching it at the theatre before Watchmen. I’ve watched it a couple more times online, it really hit the spot. Heightened anticipation for the movie.

Honda leaves F1

Honda announced at the end of last year that they would leave Formula 1 due to the the effects of the global economic climate. Now that the have found a buyer for the team, they have posted a thankyou note on their website. I’m sorry to see them go, Honda has a true motorsport pedigree. I am reminded of this advertisement that expresses the spirit of Honda.

The Honda F1 team was bought by Ross Brawn, a former technical director of Ferrari, and renamed Brawn GP.