No new report yet. Didn't have time.
you are the facto changing the plugin to limit the competitor to only who what make you look good and ignore the rest.
With all due respect, you are making assumptions, which is what you criticized in others. You are assuming that my only goal is to make PhysX look good. This is wrong.
Moreover, you do not seem to be reading what I write. I said that things like new shapes would have to wait for the next version, because I have to implement them in all engines. For a shape like a cone, implementing them in all engines
except PhysX is the
opposite of making PhysX look good. Thus, this is wrong again.
Please note that PEEL is useful to compare engine A to engine B. To do a comparison, you need to have at least 2 engines exposing the same feature. Otherwise this becomes just a sample or tech demo for Newton. And that's your job, not mine.
I am forced to tell you the same as what I already told the Bullet people who complained before you: if you are not happy with my work, feel free to create your own framework with your own tests and your own comparisons. It is probably the quickest and most efficient way to prove your points.
I also said Newton was a far more feature rich engine than any one on that list. (particles soft body stuff is not physics)
Out of curiosity, what feature do you have in mind here, that others do not have?
In any case, a feature only supported by a single engine is of little interest to me for PEEL, since, again, I would have nothing to compare it to. Obviously I cannot add to PEEL each unique feature of each supported engine.
when you added that last 255 x 255 array, you quickly jumped to celebrate how cool your stuff was because no one else did that.
It is easy to prove this wrong. Just go back to my initial blog posts from 2 years ago:
http://www.codercorner.com/blog/?p=864Even Bullet (and others) can handle that scene, so there would have been no reason to "celebrate" anything.
The Newton integration did not exist at that time (you complained loudly enough about it), so there was no way for me to know that Newton would crash here.
Let the record show that the comparison is a "Low common denomination feature set show"
I am fine with that. To do a
comparison, you need to have 2 elements indeed, i.e. some common feature set. Otherwise, again, it's just a show case for one particular engine.
Not true, It is in newton since day one and I added to the plug in I turned in.
It is in Newton, yes. But it is not in other plugins. Look for that function in your own PEEL master branch, it is only implemented in Newton. So unless I'm missing something, what I wrote was true.
That was how back then the experts falsely discredit Newton to favor ODE and Noveded, by not setting the angular damp to zero
As far as I can see this is exactly what you just did as well. You did set the angular damping to zero in Newton, while others kept using their default non-zero value. I'm sorry but from where I sit you just did exactly the same.
No true either when you pass a radio in engines like Physx, havoc and Bullet the radio is augmented by an implicit margin. Newton does not add collision margin. So I look at the default collision margin and added the extra size so the two ball are the same size.
The balls are not the same size. If you change the test in the PEEL code, the same modified radius will be passed to all engines. Newton will still not add a collision margin. The others will still add a collision margin. Thus the radius will be different anyway - no matter what initial radius you put in the test code.
But that's fine. That's exactly the point of PEEL: checking what different engines do when given the same input data. That's a comparison.
So yes the test was disingenuous.
No, it cannot be: we did not even write that test ourselves. You are again not reading what I write. The data came verbatim from a customer. It's their values from their game. We cannot change that. If they would switch their physics engine from PhysX to Newton out of "desperation", they would see exactly what PEEL is exposing here. That's the point of the tool. The tool is working.
No even joint limit, want what else, are you going to change?
You are once again not paying attention. I wrote that this addition was
good. But I need time to implement it in all engines. As for what else I am going to change, I will let you know here in my next reports.
With all due respect, you are damaging your reputation on your own forum, in front of your own users. Please stop. I am still trying to work on this with you here.
(and I have little time to devote to PEEL these days).