A lot of people have been waiting for it. It now happens.
Saul Klein started it last year with TAG, now launching the most American-like platform in Europe to help, promote and fund young entrepreneurs.
Seedcamp takes inspiration from the know famous Paul Graham's YCombinator offering what is painfully lacking here on the old continent: an attraction pole for daring young developers to gain confidence and enough money to launch their company.
This "Start-up Idol" concept can be seen by some people as very expensive for the funded company: from the 20 selected, 5 winners will receive €50K each for a 10% stake. That's not really a bargain I think, but all the mentoring and networking provided by Seedcamp is probably the most valuable part of the deal when you are a business newbie.
Even though I am not really in the "young" entrepreneur category anymore (argh!), I am going to apply with my new project (that is still secret, sorry). Hopefully I will gain a lot of insight from this experience.
Thursday, July 19, 2007
Tuesday, July 10, 2007
... and back to the ground
The peak in the Sportstar Simulator's stats did not stay for long. I still do not know what was the cause of this. Anyway, the game is still doing between 500 and 1000 hits a month and I am happy with that.
I, sometimes, receive an email from a visitor asking about the next version/improvement/bug fix and I try to do my best to make them happy. But I guess the simulator is now going to stay as it is.
The original idea was to implement a terrain engine to stream data (textures and elevation) from the server and offer a greater area to explore. That was really challenging and I gave up to be able to publish a version in time. Shockwave (Lingo) was not fast enough to build a proper ROAM algorithm and I had to come up with a homemade Level Of Detail system; just to discover that the texture swap was slow too, freezing the game for a few milliseconds every now and then. Using Havok for collisions was a nightmare when applied to a dynamic mesh and my shared hosting server was struggling to generate and deliver the textures.
I sometimes think it could be nice to use the Google map images as textures in such an application: they have the big machines and broad pipes. It's all nice and sliced up in little squared chunks, ready to be mapped on the mesh and fast delivered :)
Another time may be. For now the terrain is a single mesh object with a poor resolution but easier to integrate in the game. The focus, anyway, was on the plane and its flight characteristics.
I, sometimes, receive an email from a visitor asking about the next version/improvement/bug fix and I try to do my best to make them happy. But I guess the simulator is now going to stay as it is.
The original idea was to implement a terrain engine to stream data (textures and elevation) from the server and offer a greater area to explore. That was really challenging and I gave up to be able to publish a version in time. Shockwave (Lingo) was not fast enough to build a proper ROAM algorithm and I had to come up with a homemade Level Of Detail system; just to discover that the texture swap was slow too, freezing the game for a few milliseconds every now and then. Using Havok for collisions was a nightmare when applied to a dynamic mesh and my shared hosting server was struggling to generate and deliver the textures.
I sometimes think it could be nice to use the Google map images as textures in such an application: they have the big machines and broad pipes. It's all nice and sliced up in little squared chunks, ready to be mapped on the mesh and fast delivered :)
Another time may be. For now the terrain is a single mesh object with a poor resolution but easier to integrate in the game. The focus, anyway, was on the plane and its flight characteristics.
Subscribe to:
Posts (Atom)