Texture sizes

[ramsa329 0]

No actually Bob I'm not saying that at all at all or AT ALL.Why in the world would I?What would ever lead you to believe that.My purpose is the one I keep repeating and for some reason you keep looking for some OBSCURE REASON TO DOUBT.I want people to become more aware of what is going under the hood of the sim.and I believe that this has more importance then has been addressed up until now.PERIODP.S. This has nothing to do with me personally so if you don't mind try to keep from using me as a personal reference.

[bob.bernstein 0]

ramsa...you misunderstood...I was talking to MIke, not you with this post. You responded to the wrong post.Mike really did say those things...and so my critique stands as written.I will say that both of you have missed the boat. Your effort as you describe it above is still directed at the developers. Developers will do what earns them market share and profit. Period.You want to change what the developers produce, change what the customers are demanding and what they are deciding to pay for. Simple as that.Nothing here is helping the customer to stop demanding excellance in appearance and function...and that customer will unknowingly reward the dev that delivers an increased poly count and features...AND then will gripe about performance. But note...THE SALE OCCURRED before the gripe. Until the sale doesn't occur becuase its too good looking or has too many features..you guys are wasting your time.Bob

[Ramasurinen 0]

Bob, what I tried to say with my post is that you actually can have your cake and eat it too when it comes to performance and I'm speaking from personal experience at an extreme end of the development scale. I'm going to name names with this one because in this case I think it's important to give some credit where it's due. The only plane in my hangar that actually performs as well in multiplayer as the default aircraft is the RealAir SF260. It's a whopping 19000 polys, it's highly detailed, it looks gorgeous, and I can have *piles* of them on screen before my fps tanks. Quite simply that makes it a joy to fly in multiplayer for me, and multiplayer is where I spend 100% of my time in FSX. Am I in the majority? Nope, not by a long shot, and I'm well aware of that, but efficiency is efficiency regardless of how you use the sim. I know the singleplayer legacy is strong and lives on, but it doesn't change the fact that many aircraft out there could be a lot leaner on resource requirements without losing much if anything in visual detail. One of those 100k poly monsters I mentioned before doesn't look anywhere near as good as the Marchetti, inside or out, and yet this 100,000 poly plane in question is just a single engined, single seat, weaponless tube with wings really. The only point I was really trying to make with my post is that texture efficiency is only part of the performance problem, the model's efficiency still counts for a lot. Efficiency does not have to come at the expense of detail however, it just takes a little more planning is all.What my years of lower poly modeling experience has taught me is that you can get away with a *lot* of texture tricks if you put the time into efficient modeling of whatever it is that you're making. The effect Ramsa is seeing by reducing texture sizes and so forth is a cumulative effect of reduced render time per poly, and I'm guessing the planes he's referring to have quite a lot of them. Ripping off bump maps is something I have to do quite often just to get some of my planes to perform decently in multi, and I have a blisteringly fast system.As for sales occurring or not, I do vote with my wallet, and there are some developers I probably won't purchase from again, and some that I know I can buy from without worry. Unfortunately there's no way to know the damage a plane does performance wise until you purchase it. You can at most only make an educated guess based on the general complexity of the real aircraft being replicated and on the developer's previous efforts. Developers don't publish poly counts and reviews rarely if ever cover multiplayer performance.Unfortunately for myself and other multiplayer fans that makes things a bit of a crapshoot, and though we're a relatively small voice in a big singleplayer ocean our performance suffers the most.-mike

[ramsa329 0]

I apologize if I reponded to the wrong post it happens that my name is also Michael.However if I were Mike (which I am so it even more confuses "old" men like me) I'd still respond to you the same way since I don't believe that he was saying that at all nor do I see how you arrived at that reading his post.Seems a bit too literal a reading.The issues you are raising have a lot of legitimacy and I would be the last to deny that.However, they are bordering on the "chicken and the egg".Consumer demand vs. developer responsibility or your or my support of either side is not the issue here.Consumer and developer AWARENESS is however the issue.Whether the consumer desires higher and higher amounts of "detal" it is still up to the developer and their testers to insure that it is delivered at the best possible levels of function.and this post is about that and nothing more.If you feel the post is a waste of time and space then request that it be stopped since you seem very convinced that it will do no good and have repeated that over and over.If it is stopped then fine. If only one person opens a texture folder of a plane that doesn't seem "quite right" and starts questioning what's going on then we have served a purpose.AWARENESS

[bob.bernstein 0]

<>THis is the crux of the debate...the developers would say they have delivered what the customers reward "at the best possible levels of function".No one intentionally delivers less efficient output than they are capable of, in my opinion.Ramses, I see these discussions as a positive, when I say "waste of time" I don't mean the discussion is a waste of time, I mean you are not making progress toward your goal until you attempt to change the customers desires. Only then will you enjoy results for your efforts. Its not chicken and egg at all...its all one way, customer makes the decision what to reward, developer sees what the market wants and delivers.

[ramsa329 0]

The assumption I think you are making is that these problems (if we assume that there are)are being generated by customer demand. This implies that the designers must improve detail by higher and higher standards due to this demand which I do not disagree with at all.Where I disagree is in the assumption that the detail must just by the fact that it is being incresed cause problems.It is at this point where we part company. I strongly believe that detail can be increased at the same time and in balance with performance and visuals.I do not see the 3 as being mutually exclusive at all.However when I open a texture folder and see textures of 16000KBs individually, I cannot help but wonder and would find it strange if you wouldn't also.and I doubt that the consumer asked for this.So back to my original issue is it getting out of hand or not?

[bob.bernstein 0]

Any single peice of evidence is not sufficient to determine a trend. Things "getting out of hand" is a form of claiming a trend exists.That one developer may have made an error of judgement, however if that judgement was that developer's best effort, nothing discussed here would alter that fact.The idea that developers (competitors) would help each other avoid looking dumb by sharing best practices is unlikely except at a very general level as Bill began to do in the beginning of this thread.So, this boils it all down, you assert that developers "should" be capable of meeting customer demand in the way you deem "wise", yet if they don't know what "wise" is, they still won't acheive it.Finally, no matter how wise a developer is capable of, if he has a legacy product that is less than "wise", he/she may still release it as a business decision. In that case, the rework may be not economical, and the release will generate additional income.Anyway, we've probably never been discussing anything with practical application because each idea you've recommended will result in lower incomes for developers than current practice and you've offered no incentive for any individual developer to incur those losses.The only logical thing to discuss is to share best technical practices, which Bill Leaming began to do, and the OP or perhaps it was you...someone claimed that was not the mission of this thread. Yet the only way a developer can change anything for free is to to employ new ideas on the next project, which only comes from technical interchange on the theorectical level. Cheers,Bob

[ramsa329 0]

Bob,There will never be an end to this.I disagree to the highest extent with just about everything you say and you likewise disagree with me.I find your observations about income loss to be ridiculous and your assumptions about other posts that have been included here to be partial reads at best.In fact actually I think you didn't even fully read or understand them.As to Bill I am sure that had he desired to continue with whatever he was saying and felt it was important he could have started his own thread.So in order for this back and forth nonsense to stop and since I have too many clients who need my help and am way too busy trying to earn a living.I CAPITULATE and BOW to the higher order.Thread over.Yours,

[Guest -BeNt-]

Bob, I've been reading this entire thread for a few days now. And I have found it very interesting on both sides of this conversation. But to clarify, merely from my understanding here, is that you do need to convince the developers to develop a high performance plane instead of a bloated plane, not because they are being "rewarded" because that is what customers want it's because that what customers have come to expect. In the 60's car buyers expected a big car with a big engine that could go fast, and if you told them in 90's that they would be driving little cars with little engines that could go fast they would have laughed at you.The customers, whom most, don't have a clue that the performance of the sim can be so much better if the plane is designed, built, and textured more efficiently have come to the point that they blame the sim, hardware, and whatever else they can find to be the excuse the sim is running bad. But when looking deeper into things it could be a vast thing between scenery and planes not being built as efficiently because devs. believe that people want higher poly counts etc. And in one way you are 100% correct up until that customer gets those things, and then comes here complaining because they can't get the bird bird flying right, or smooth/good framerates coming into their new sceneries.And the comment of customers demanding fewer features and less detail is a horrible backhanded comment. When in reality what these guys are saying is you can get the same amount of detail and features if everything is carefully planes, faithfully executed, and un-doubtfully modeled, textured, and tested to perfection instead of rushed to scene because that's "what customers want", which in my opinion is absolutely untrue considering I want performance and features instead of having to tweak just about every aircraft I get to gain the performance it should have had from the start. And market demands don't drive all products in this community. Most if not all of some popularity of planes pull off of hype and mis-information which ramsa eluded to in an earlier post. Since most simmer's don't know of the performance the popular plane they have that almost flys right, can have once tweaked right(I'm not talking config tweak either) then they just assume that's how it runs because of the sim, and not because of the airplane and come back here bragging about how great the plane is, and without starting a flame war, the over bloated 747 is an example. Horrible, absolutely horrible.

[bob.bernstein 0]

<>And I'm saying these guys aren't being scientific and may not know what they are talking about. The comparison of "goodness" is the default. See Mike T's post. Comparing goodness to the default means either1. The default aircraft were not produced as efficiently as possible or2. reduced drain on the sim is deemed appropriate in exchange for more features and polygons.Can't be anything else. Which is it?BobAnd if its the second, how much is too much...good luck defining that.

[bob.bernstein 0]

Based on the Real Air plane example, perhaps that's the data people are using to say that other "can" get there.Are you saying they "must" be as good as Real Air or they shouldn't develop product? Every marketplace has excellant, medium, and poor performers.The market sorts out good from bad. Poor performers fail, unless they improve. Medium performers are probably on the path to becoming excellant. Excellant performers didn't start out excellant.So, this thread seems to be saying....guys....don't be on a continuum...don't start out weak and grow strong....no...everybody just be excellant right now! Don't require learning to get there.In a world of free enterprise, this type of advise is market based, not thread based on Avsim. Since the beginning there have been products of varying quality. Buy the good, don't by the bad.

[virtualstuff 759]

>Bob,>>There will never be an end to this.>>I disagree to the highest extent with just about everything>you say and you likewise disagree with me.>>I find your observations about income loss to be ridiculous>and your assumptions about other posts that have been included>here to be partial reads at best.>>In fact actually I think you didn't even fully read or>understand them.>>As to Bill I am sure that had he desired to continue with>whatever he was saying and felt it was important he could have>started his own thread.>>So in order for this back and forth nonsense to stop and since>I have too many clients who need my help and am way too busy>trying to earn a living.>>I CAPITULATE and BOW to the higher order.>>Thread over.>>Yours,>> Funny this thread it's real simple actually in 3ds max make sure you watch your vertices (because they count but hey vertices are part of the polygon and watch carefully the face that's count ;-)As for the draw calls when you are smart in 3ds max you can combine a lot into a single draw call regarding materials ;-)But then again I don't know anything about GMAX only 3ds max :-lolhttp://i5.photobucket.com/albums/y156/awf1/sign.jpg

[n4gix 4,947]

>But then again I don't know anything about GMAX only 3ds max>:-lolFor all practical purposes, they are the same. GMax is simply Max v3.5 lite... ;)Optimization of a model is not a simple process. It is the end result of dozens of design decisions working in concert to produce the best results with the least impact.As it happens, there isn't any "one" technique that's best for every project... :-lol

[n4gix 4,947]

>As to Bill I am sure that had he desired to continue with>whatever he was saying and felt it was important he could have>started his own thread.While it's true I could have done that, what would be the point? I was attempting to answer what turned out to be front-loaded "questions." Perhaps by not being smart enough to recognize it right away, I honestly invested valuable time and tried to provide useful, accurate information...There's little to be gained by speaking into a vacuum.

[N2056 33]

Well, I'd have to say that your efforts (and a few others) have not been in vain, as there have been a few things I learned reading this thread.