Bottom to top Approach- video/ graphic to Developers
Marksman-Specialist
Join Date: 2011-01-18 Member: 78049Members
Join Date: 2011-01-18 Member: 78049Members
Don't flame me if my interpretation is wrong. I know that gaming now day requires almost extreme specification just to even play a games. But I know graphics alone isn't everything without good gameplay function. I know many/ majority of computer users have at least a duo core processor and 3-4gb of ram, but lacks good video card. Not everyone can afford high end computer that last for the next 5-6 years.
I know that the Devs are trying to put out a quality product, and trying to keep up to day with the technology and graphics. However, I not sure if its called a Top to bottom approach, because the engine right now is optimized to work with the most intense graphic first, thus leaving many of the beta testers out of reach to even run NS2. Less beta tester leads to slower development. Thus customers are unable to truly participate in the beta and help clean up the bugs.
Maybe the Dev could start from the Bottom to top Approach instead, start with not so CPU intensive specification first and then increase the graphic/ shaders/ pixel, etc by increments per patch in the settings. this way more people can participate and work more efficiently and produce polished product.
I know that the Devs are trying to put out a quality product, and trying to keep up to day with the technology and graphics. However, I not sure if its called a Top to bottom approach, because the engine right now is optimized to work with the most intense graphic first, thus leaving many of the beta testers out of reach to even run NS2. Less beta tester leads to slower development. Thus customers are unable to truly participate in the beta and help clean up the bugs.
Maybe the Dev could start from the Bottom to top Approach instead, start with not so CPU intensive specification first and then increase the graphic/ shaders/ pixel, etc by increments per patch in the settings. this way more people can participate and work more efficiently and produce polished product.
Comments
Firstly you get your game working with all the required systems, then you optimize optimize optimize there is no point doing to much optimizing upfront because by the time you have done a feature freeze much of the code may have changed and you will need to optimize again.
Its like putting spices in the oil of your pan before adding all your ingredients -> it doesn't work.
You put in your ingredients let it cook a bit then flavor it roughly with your spices, cook a bit more, maybe add some more ingredients and slowly optimize the flavor to your desired taste.
Atm we are cooking and flavored it roughly - we can already try it... it doesnt taste perfect (some maybe cant even stand the taste yet, the are not hungry enough to eat it, or their spoon is too tiny to have the whole flavor in one bite) but we know it will get better the more ingredients and spices are added.
oO wtf am i writing :X does this make sense?
Please let us know which games you've developed - I'd like to give them a look.
But your interpretation is wrong.
In SOME areas developing software in the manner you proposed occurs.
These areas are predominately related to systems where bodily harm can occur.
Then a phased infrastructure approach can occur.
In MOST areas developing software starts with a list.
The list is implemented, Then the product is optimized.
Sometimes optimization includes chopping items off the list.
UW have been performing some steps at optimization while also chipping away at their list.
Each build has shown progress and the game has become more playable...and FUN.
While playing you should enable the fps stats and the network stats.
This should give you an idea of whether the problems you are seeing are related to rendering (fps) or multiplayer (network) play.
Another test would be to start a server on your machine and join it.
Just this information will help UW make a better game.
You can report it or simply add you name to the list of existing problems.
<a href="http://getsatisfaction.com/unknownworlds/products/unknownworlds_natural_selection_2" target="_blank">http://getsatisfaction.com/unknownworlds/p...ral_selection_2</a>
or you can complain about it in the forums.
UW is very aware of the performance problems and working on them.
If you check this page
<a href="http://www.unknownworlds.com/ns2/progress" target="_blank">http://www.unknownworlds.com/ns2/progress</a>
Both these tasks are optimization related.
Profile and optimize OnProcessMove()
Hydra optimization (create 5+)
"Premature optimization is the root of all evil"
One of my CS finals required an explanation of this quote. :)
WHAT? No. That's actually an excellent way to cook. You flavour the oil (and in some cases roast the spices to release the oils and aromas), then the flavoured oil in turn flavours the meat and vegetables. Cooking it the other way is for seasoning the surface, like adding pepper on top of your eggs.
Also, basil.
I like to stir-fry green beans with a bit of olive oil, some fresh crushed garlic, a bit of salt and lots of pepper.
edit: i knew i should have deleted it. [At least its for the luls.]
I know that the Devs are trying to put out a quality product, and trying to keep up to day with the technology and graphics. However, I not sure if its called a Top to bottom approach, because the engine right now is optimized to work with the most intense graphic first, thus leaving many of the beta testers out of reach to even run NS2. Less beta tester leads to slower development. Thus customers are unable to truly participate in the beta and help clean up the bugs.
Maybe the Dev could start from the Bottom to top Approach instead, start with not so CPU intensive specification first and then increase the graphic/ shaders/ pixel, etc by increments per patch in the settings. this way more people can participate and work more efficiently and produce polished product.<!--QuoteEnd--></div><!--QuoteEEnd-->
as it is the engine isn't optimized at all (neither for high or low end) - thats what they are working on now, optimizing the hell out of it
That's not true. Some identifiable/visible optimizations are occlusion culling and simple graphic settings. Occlusion culling is probably the biggest optimization. There are blog posts that state that they already did a bunch of optimizing.
I agree! BUT don't put chilli powder in the pan with just oil unless you enjoy choking on burnt chilli fumes... chopped chilli's is probably fine.
Now the fight about what Unknown Worlds is doing.
Ive quite recently found out how much better food tastes after it seasons with chili and black pepper and cream inside a fridge overnight.
[EDIT:] Duh, I forgot Garlic, onions and cheese. That lot goes well with beef.
related to CPU being utilized properly with multiple cores alongside LUA garbage collector and onprocessmove()
or at least... thats what i've put together myself?
Apart from that, nothing wrong GPU wise so far.
The actual quote, by Donald Knuth, is:
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->"We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil"<!--QuoteEnd--></div><!--QuoteEEnd-->