Annoying Counter Errors in Populations

buzzzzz1 · May 03, 2009, 02:04:09 PM

Is anyone else getting these annoying population counter errors as constantly as I am? Or has this already been discussed and is there a solution or a fix coming out? What happened to the Alpha Testers? Are they still testing? I can't believe that planetside released what's supposed to be a Final release with all these problems I've seen mentioned here. And on top of that all the Bull Shit that we were fed by Xfrog. It's just a bit much! Grrrrrrrrrrr

dandelO · May 03, 2009, 02:13:47 PM

I do.

The thing is, though, I seem to get them only during low detail test renders. For example, I have a scene with 2 populations - one grass, one trees. Test renders above 0.3 detail give no errors, below this(at 0.1 or so for simple testing and placement purposes) throws out an error of something like: Counter error in populator/pop grass clump
The render always looks fine, even with the error, no missing patches of grass in view.

This is with built in TG grass object, I've only seen it with this though, never an imported object.

buzzzzz1 · May 03, 2009, 02:20:58 PM

It doesn't seem to matter what the quality or the size of the render for me. I'm working on a scene with 6 imported populations and 8 single imported objects. Task manager is showing 854 MB of mem usage and 1604 MB VM size. So I don't think it's a memory problem because I'm running XP64bit with 8 Gig of Ram.

neuspadrin · May 03, 2009, 02:42:41 PM

I've noticed this since full release. Don't really remember it from before.

Jack · May 03, 2009, 02:56:28 PM

i get this 2 i have a scene that I am not able to render because of it

Tangled-Universe · May 03, 2009, 03:41:45 PM

I have this virtually always and it is nothing to really worry about, though it is annoying.
It is part of a warning system that it encountered an error, but it will not affect the final render.

Martin

rcallicotte · May 03, 2009, 06:46:47 PM

Martin is right. I remember Oshyan mentioning this is only a warning. Not sure if it's related to the bigger BLACK BOX errors or not.

Tangled-Universe · May 03, 2009, 06:54:13 PM

Quote from: calico on May 03, 2009, 06:46:47 PM
Martin is right. I remember Oshyan mentioning this is only a warning. Not sure if it's related to the bigger BLACK BOX errors or not.

No, those errors were introduced in the most recent versions and are probably related to some memory-management problems. Reducing cache-size often solves the problem.
The counter errors are from versions starting around ~september last year (remember my brand new day image couldn't benefit then from the later fixes).

FrankB · May 03, 2009, 07:02:06 PM

Quote from: buzzzzz1 on May 03, 2009, 02:04:09 PM
Is anyone else getting these annoying population counter errors as constantly as I am? Or has this already been discussed and is there a solution or a fix coming out? What happened to the Alpha Testers? Are they still testing? I can't believe that planetside released what's supposed to be a Final release with all these problems I've seen mentioned here. And on top of that all the Bull Shit that we were fed by Xfrog. It's just a bit much! Grrrrrrrrrrr

Hey Jay, we don't have our Mr. Nice Guys shoes on today, do we?

Those counter errors are knows to PS , of course. Because they usually don't make any troubles, they're probably not very high on the to do list - which as you know - is being dealt with sort of sequentially, because there's only one programmer of the renderer.

I believe I do see a pattern that is related to certain objects. Not that I have done any scientific research on it, but I have this mostly with (very many) single grass objects. But that's all guesswork, I can't really put my finger on it.

Cheers;
Frank

buzzzzz1 · May 03, 2009, 11:20:11 PM

Sorry, I haven't been to my anger management classes lately nor have I had my Bran Muffin today.

Oshyan · May 04, 2009, 12:57:29 AM

As Frank mentioned, we're aware of these problems, and they've been around for quite some time. There seems to be a tendency to blame the final release(s) for errors now occurring, as if they were freshly introduced, but as far as render-specific problems go, I've yet to see anything that is really new to these last few releases. This is probably because the renderer has not undergone many major changes since the beta release.

In any case as several others have said, the counter errors usually do not affect the final image, so you don't generally need to worry about them. The counter errors are quite honestly a bit elusive and difficult to track to their source, but they're definitely something we're working on. Fortunately they tend not to have a noticeable negative effect.

- Oshyan

Tangled-Universe · May 04, 2009, 03:55:38 AM

Quote from: FrankB on May 03, 2009, 07:02:06 PM

...

I believe I do see a pattern that is related to certain objects. Not that I have done any scientific research on it, but I have this mostly with (very many) single grass objects. But that's all guesswork, I can't really put my finger on it.

Cheers;
Frank

Is that because you primarily use grass objects over trees? In my experience it doesn't matter which kind of model or the amount of instances.
In my jungle-image I had 1 million grass instances and they gave no counter errors, but the "couple of" (50k) oaks I had gave countless errors.
In my experience it is pretty random / unpredictable.

Martin

buzzzzz1 · May 04, 2009, 09:03:26 AM

Quote from: Oshyan on May 04, 2009, 12:57:29 AM
As Frank mentioned, we're aware of these problems, and they've been around for quite some time. There seems to be a tendency to blame the final release(s) for errors now occurring, as if they were freshly introduced, but as far as render-specific problems go, I've yet to see anything that is really new to these last few releases. This is probably because the renderer has not undergone many major changes since the beta release.

In any case as several others have said, the counter errors usually do not affect the final image, so you don't generally need to worry about them. The counter errors are quite honestly a bit elusive and difficult to track to their source, but they're definitely something we're working on. Fortunately they tend not to have a noticeable negative effect.

- Oshyan

OK?, I guess I'll never understand how you can release something and call it a Final Release with known issues. Perhaps someone can educate me in software terms.

Thanks for all the replies,

Oshyan · May 04, 2009, 11:00:16 PM

Virtually every software release in history has had "known issues". In the software industry "perfection" is all but unattainable. The best a developer can hope to do is eliminate the majority of issues that regular users encounter. But inevitably there will be those problems that even 1 user encounters. That's a best-case scenario, most of the time, even with large professional software publishers, there are many known issues in every release, and even more that are *not* known. Look around at forums for 3DS Max, Maya, Mudbox, Vue, etc. and you'll see plenty of similar issue-related discussion.

- Oshyan

Walli · May 05, 2009, 11:17:30 AM

Quote from: buzzzzz1 on May 04, 2009, 09:03:26 AM
OK?, I guess I'll never understand how you can release something and call it a Final Release with known issues.

For sure I am not able to educate you ;-)
But I am 150% sure that NONE of your applications that you have installed comes without issues. If you haven´t had issues, then this usually means that you simply were more lucky then other users ;-)

Annoying Counter Errors in Populations

dandelO