Can anyone give me the acceptable range for poly counts for a weapon in a fps type game today? I’d just like to know so I can keep it in mind as I work on portfolio pieces
The answer is always the same: As few as possible and as many as necessary. This can be a single face, a simple billboard. This can be a million polys when you plan to dive into the surface scratches of your weapon. It all depends of your needs. No game is equal. And polys isn't the showstopper anymore. It's shaders, materials, particle fx, level size and number of assets, game logic code and so on. And it depends of what target platform you choose. A smartphone game can't handle geometry like a high end pc with the latest gamer gear.
For a first rough value have a look at games similar to what you want to develop. The best method is to make a prototype or a little test scene though. And simply test it out. When your weapon looks too blocky then it most probably needs more geometry / higher texture resolution. And when the fps goes down to unplayable then you need to optimize somewhere. When this happens by a single weapon then you have another problem though. But this shouldn't bother you for pure portfolio work. Here i would make it best looking.
For portfolio stuff you should be more concerned about clean topology, which should flow naturally to efficient polygon use. Orient your thinking around whether your edges/faces/vertices contribute meaningfully to the models silhouette, or whether they could be replaced with baked details.
Try to focus on the visual quality of your work before its technical aspects. It's very easy to fixate on numbers like polygon counts (or draw calls or texture size and so on and so on) and your work suffers for it.
All that said, I usually fit a very rough benchmark for 30k tris. When collecting references for your model, look at what other artists have done for similar pieces - they will often post wireframes and/or polygon counts.
Replies
Hey msegarra,
The answer is always the same: As few as possible and as many as necessary. This can be a single face, a simple billboard. This can be a million polys when you plan to dive into the surface scratches of your weapon. It all depends of your needs. No game is equal. And polys isn't the showstopper anymore. It's shaders, materials, particle fx, level size and number of assets, game logic code and so on. And it depends of what target platform you choose. A smartphone game can't handle geometry like a high end pc with the latest gamer gear.
For a first rough value have a look at games similar to what you want to develop. The best method is to make a prototype or a little test scene though. And simply test it out. When your weapon looks too blocky then it most probably needs more geometry / higher texture resolution. And when the fps goes down to unplayable then you need to optimize somewhere. When this happens by a single weapon then you have another problem though. But this shouldn't bother you for pure portfolio work. Here i would make it best looking.
Kind regards
Tiles
For portfolio stuff you should be more concerned about clean topology, which should flow naturally to efficient polygon use. Orient your thinking around whether your edges/faces/vertices contribute meaningfully to the models silhouette, or whether they could be replaced with baked details.
Try to focus on the visual quality of your work before its technical aspects. It's very easy to fixate on numbers like polygon counts (or draw calls or texture size and so on and so on) and your work suffers for it.
All that said, I usually fit a very rough benchmark for 30k tris. When collecting references for your model, look at what other artists have done for similar pieces - they will often post wireframes and/or polygon counts.
Just check the stuff on Artstation there is a lot to find.
https://www.artstation.com/artwork/8l8LyE
Okay thanks that’s what I’ve been learning to shoot for anyway the clean topology.