Re: [Plib-devel] Re: KobayashiMaru & plib

[Wolfram K. <w_...@rz...>]

Steve wrote:

>I'm conscious that I always underestimate times for things.  

I think almost all programmer do this. The difference is between the
programmers who know this and who know their factor and those that
don't. Also, there are certain things you can do to get a more
accurate estimate. One danger is that you know all the small steps of
the next problem - but don't think about all the small steps of
problems later on.

When I estimate something as one day, then I can be wrong by up to
100%. When I estimate something as, say 2 month then I am normally
pretty accurate. Of course some things are more difficult to guess.
For example, estimating the implementation a algorithm that is in my
head is normally fairly accurate. But if I have to use a new API 
or file format I have never used before, then its much harder.
So, besides giving an estimate I can foretell the accuracy of my
estimate.

>>I didn't find it, maybe you will find it in the
>> archives.
>
>It might have been on the TuxAQFH or TuxKart lists - I don't remember.

No, I am not subscribed to those lists. It has to be one of the lists
I mentioned.

>> And Steve didn't say the problem was that terragear is unsuitable. He
>> said that to get good frame rates you have to tweak the game design,
>> that is, you cant have a universal game engine with todays hardware.
>
>...or ANY hardware EVER.  The problem is that people's expectations
>have gone up as the hardware gets better.

Since I am fairly conservative in what I say, I wouldn't say that. But
certainly not for a long time. BTW, I just wrote about the growing
expectations on slashdot a few days ago. I think there is still quite
a lot of room for growing expectations.

Bye bye,
Wolfram.