<!--quoteo(post=1727763:date=Sep 18 2009, 03:04 AM:name=Sirot)--><div class='quotetop'>QUOTE (Sirot @ Sep 18 2009, 03:04 AM) <a href="index.php?act=findpost&pid=1727763"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->It's not really easy. See my earlier posts regarding it being difficult.<!--QuoteEnd--></div><!--QuoteEEnd-->
Right, middleware, like the stuff used in rainbow six would be difficult to implement, but how about if every game directory has a pre-made folder where you keep your various face-scans images. You just put the image files of your face in the folder, and they appear ingame. This option to import image files as opposed to connecting the game to your hardware would be actually very easy, similarly easy to how valve games allow you to import custom sprays
however, the lack of forcing a webcam device to take the picture will motivate some [webcam-less] people to just import pictures from their computer or from the internet rather than actually scanning their own face. But I dont really see this as much of an issue considering it could be easily countered by 1) a server option to disable custom imported faces 2) an option to disable the imported face of a particular player in the server (similar to a server mute function) and 3) an option to not be able to view any custom faces at all.
All together, I agree with the topic creator that this would be a capitol Idea, and it SHOULD indeed be implemented; it should not be at ALL difficult or time-devouring to add to the game. And I think it would really add a nice flare to the game. I fully support this idea considering how delightful it would be to see the faces of the people your fighting with/killing you with mixed reactions.
<!--quoteo(post=1727947:date=Sep 19 2009, 02:22 AM:name=Sirot)--><div class='quotetop'>QUOTE (Sirot @ Sep 19 2009, 02:22 AM) <a href="index.php?act=findpost&pid=1727947"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Give an initial selection of like 5 or 6 faces then offer face packs as a way to raise money among the devoted playerbase.<!--QuoteEnd--></div><!--QuoteEEnd-->
haha, no
<!--quoteo(post=1727940:date=Sep 19 2009, 12:50 AM:name=slayer20)--><div class='quotetop'>QUOTE (slayer20 @ Sep 19 2009, 12:50 AM) <a href="index.php?act=findpost&pid=1727940"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Yeah, I don't want servers to have the ability to force players to download skins like faces.<!--QuoteEnd--></div><!--QuoteEEnd-->
Think of it this way, when a new player enters a game server like in counterstrike or team fortress 2, you have to download the skin of their spray, but it's not like this is a huge file to download. Games have done this sort of thing before, come on, if your so concerned about this lag issue then the skin should be one of the first things downloaded to the server, and set to download at a certain rate.
I for one do not forsee any sort of lag problem if there is only 1 face per client, and it isn't to large of a picture. Also, consider that Arma is a game with MANY people per server, and ALSO consider that only half the team are going to be applicable for these custom faces.
<!--quoteo(post=1727781:date=Sep 18 2009, 04:45 AM:name=Sirot)--><div class='quotetop'>QUOTE (Sirot @ Sep 18 2009, 04:45 AM) <a href="index.php?act=findpost&pid=1727781"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I will say right now that I never studied or programmed any software or application that any way involved scanning a face and creating a 3d model using its features. That said, the knowledge that I do have tell me it's a bad idea. There are two ways of doing something like this:
VARIANT A: The picture is scanned, facial features are found and a model is built around those features. The face is used as a texture for the created model. VARIANT B: The picture is scanned, facial features are found and sliders move around to construct a similar face to the one pictured. The face's texture is created from stock materials.
1. It will increase minimum requirements of the game. Variant A will increase it drastically because creating that model dynamically eats processing power upfront and more memory will need to be allocated to hold the face texture. Variant B will be less resource intensive but the same problems will exist albeit in a slightly smaller capacity. Even if there is no face scanning at all in Variant B, it will still increase the requirements. 2. One stray patent could end UWE. There are so many steps to this process and they are so specific that any patent holder (or troll) could easily have a solid case against NS2 because they are doing a very, very specific task. This much more of an issue with Variant A than B, but then again, face scanning is a bit of a landfield in regards to patents. 3. It is difficult to do and WILL require a significant time investment to do so. They would need to create many proprietary technologies (face scanning, dynamic building, dynamic texture) to accomplish this and Max could better spend him time elsewhere, perhaps making the game itself better. Give bullet point 1, the issue of patents makes it wiser to just get middleware and that costs money too, which is in low supply. 4. VARIANT A ONLY: I don't know if any of you noticed, but NS2 has a distinct art style and it would be need to be properly represented for the faces not to look like they don't belong at all in the game. This means that the faces would need to be built very, very specifically to avoid that issue and that again, requires a lot of time and testing to produce something presentable. 5. VARIANT B ONLY: All the facial variations are going to cost money because that different nose, that jaw and that ear are going to require a model to be created. Given how different people's faces can be, it will require a lot of little models created to satisfy the diversity of the player base.
This is a much bigger suggestion than you realize. They should just add a loadout where you can pick a face from pre-created selection and bam, you have that face. I would rather have their time spent optimizing their technology and building features that add substance to gameplay.
whoa, didn't notice this one, hehe. *cracks knuckles*
*ahem* importing models: nah
why would we import models if you can just get the next best thing by importing the texture on top of a stock face model. There would be no model change at all, simply a re-texture. If all marine models have an option to have their face texture over-layed, then this would have the memory requirements of simply importing a texture from a connecting client. This would be simple to implement, and function just a simply.
Now you might be thinking, AWWWWWWW no variant in face models? then wont my textures seem out of place on top of the model? short answer: no. There is probably already a plan to add variants to the stock marine face models. Before overlaying your imported texture, the player could decide which model to overlay their texture onto. Of course this wouldnt be as unlimited as your hypothetical in my quote, but this would be the next best thing. I dont think anyone would complain if there were limited face models to lay their textures on because JUST the FEATURE ALONE of importing your face texture so others can see would be immensely satisfying to all parties in the server.
Another concern might be, what if the stock face model doesn't match up with the face texture, well this could simply helped by having an example face texture showing where all the parts of the face should be, to help people align their face texture to the face model.
I dont know why you wouldn't support this amazing idea, and if any devs are reading this, please, PLEASE consider this. I KNOW players would REALLY enjoy this feature. As I explained many times, this would not be very hard to implement.
edit: and when I mean stock facial variants, I dont mean replaceable parts of the face like the chin/ear/whatever, I mean entirely different and separate pre-made face models, all of which with varying facial features. We dont need interchangeable parts that bad.
Its about as complicated as custom sprays in half-life. UWE's engine is reported to have revolutionary infrastructure to allow on-the-fly downloading of game code to clients. With this already in place it should be trivial to implement.
If quality control is really an issue they could simply allow players to print logo's on the back of marine armor, or something "clean" along those lines.
<!--quoteo(post=1728174:date=Sep 21 2009, 05:06 AM:name=homicide)--><div class='quotetop'>QUOTE (homicide @ Sep 21 2009, 05:06 AM) <a href="index.php?act=findpost&pid=1728174"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->This really wouldn't be hard to implement.
Its about as complicated as custom sprays in half-life. UWE's engine is reported to have revolutionary infrastructure to allow on-the-fly downloading of game code to clients. With this already in place it should be trivial to implement.
If quality control is really an issue they could simply allow players to print logo's on the back of marine armor, or something "clean" along those lines.<!--QuoteEnd--></div><!--QuoteEEnd-->
Quality Control is a huge issue for me. I would like the immersion of having the game's graphics be consistent and there not be a huge drop in quality when I look at a marine's face. Would it be a nice feature down the road? Sure, but I do consider it a huge liability unless done with a amount of care that I don't think UWE has resources to do.
If the issue is that clone marines are an issue, than ideally there should be 16 variations of marine and players get assigned one randomly for the entire match. This lets them build a presence in the game for the duration of that round, but it does have the problem that 16 variations of marine would need to be created (e.g. white male 1, white male 2, black male 1, black male 2, white female 1, white female 2...etc) and that costs money.
@N3MES1S: Just placing a texture over a pre-existing model would just in result in incredibly ugly marines.
<!--quoteo(post=1728199:date=Sep 21 2009, 02:06 PM:name=Sirot)--><div class='quotetop'>QUOTE (Sirot @ Sep 21 2009, 02:06 PM) <a href="index.php?act=findpost&pid=1728199"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Quality Control is a huge issue for me. I would like the immersion of having the game's graphics be consistent and there not be a huge drop in quality when I look at a marine's face. Would it be a nice feature down the road? Sure, but I do consider it a huge liability unless done with a amount of care that I don't think UWE has resources to do.
If the issue is that clone marines are an issue, than ideally there should be 16 variations of marine and players get assigned one randomly for the entire match. This lets them build a presence in the game for the duration of that round, but it does have the problem that 16 variations of marine would need to be created (e.g. white male 1, white male 2, black male 1, black male 2, white female 1, white female 2...etc) and that costs money.
@N3MES1S: Just placing a texture over a pre-existing model would just in result in incredibly ugly marines.<!--QuoteEnd--></div><!--QuoteEEnd-->
Yes, everyone has different opinions. This is why there are custom skin AND pure servers in team fortress 2/counter strike.
Obviously you would prefer a pure server to pertain your romantic attachment to being immersed into the game. I however will be playing this game for the lolz.
Comments
Right, middleware, like the stuff used in rainbow six would be difficult to implement, but how about if every game directory has a pre-made folder where you keep your various face-scans images. You just put the image files of your face in the folder, and they appear ingame. This option to import image files as opposed to connecting the game to your hardware would be actually very easy, similarly easy to how valve games allow you to import custom sprays
however, the lack of forcing a webcam device to take the picture will motivate some [webcam-less] people to just import pictures from their computer or from the internet rather than actually scanning their own face. But I dont really see this as much of an issue considering it could be easily countered by 1) a server option to disable custom imported faces 2) an option to disable the imported face of a particular player in the server (similar to a server mute function) and 3) an option to not be able to view any custom faces at all.
All together, I agree with the topic creator that this would be a capitol Idea, and it SHOULD indeed be implemented; it should not be at ALL difficult or time-devouring to add to the game. And I think it would really add a nice flare to the game. I fully support this idea considering how delightful it would be to see the faces of the people your fighting with/killing you with mixed reactions.
haha, no
<!--quoteo(post=1727940:date=Sep 19 2009, 12:50 AM:name=slayer20)--><div class='quotetop'>QUOTE (slayer20 @ Sep 19 2009, 12:50 AM) <a href="index.php?act=findpost&pid=1727940"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Yeah, I don't want servers to have the ability to force players to download skins like faces.<!--QuoteEnd--></div><!--QuoteEEnd-->
Think of it this way, when a new player enters a game server like in counterstrike or team fortress 2, you have to download the skin of their spray, but it's not like this is a huge file to download. Games have done this sort of thing before, come on, if your so concerned about this lag issue then the skin should be one of the first things downloaded to the server, and set to download at a certain rate.
I for one do not forsee any sort of lag problem if there is only 1 face per client, and it isn't to large of a picture. Also, consider that Arma is a game with MANY people per server, and ALSO consider that only half the team are going to be applicable for these custom faces.
VARIANT A: The picture is scanned, facial features are found and a model is built around those features. The face is used as a texture for the created model.
VARIANT B: The picture is scanned, facial features are found and sliders move around to construct a similar face to the one pictured. The face's texture is created from stock materials.
1. It will increase minimum requirements of the game. Variant A will increase it drastically because creating that model dynamically eats processing power upfront and more memory will need to be allocated to hold the face texture. Variant B will be less resource intensive but the same problems will exist albeit in a slightly smaller capacity. Even if there is no face scanning at all in Variant B, it will still increase the requirements.
2. One stray patent could end UWE. There are so many steps to this process and they are so specific that any patent holder (or troll) could easily have a solid case against NS2 because they are doing a very, very specific task. This much more of an issue with Variant A than B, but then again, face scanning is a bit of a landfield in regards to patents.
3. It is difficult to do and WILL require a significant time investment to do so. They would need to create many proprietary technologies (face scanning, dynamic building, dynamic texture) to accomplish this and Max could better spend him time elsewhere, perhaps making the game itself better. Give bullet point 1, the issue of patents makes it wiser to just get middleware and that costs money too, which is in low supply.
4. VARIANT A ONLY: I don't know if any of you noticed, but NS2 has a distinct art style and it would be need to be properly represented for the faces not to look like they don't belong at all in the game. This means that the faces would need to be built very, very specifically to avoid that issue and that again, requires a lot of time and testing to produce something presentable.
5. VARIANT B ONLY: All the facial variations are going to cost money because that different nose, that jaw and that ear are going to require a model to be created. Given how different people's faces can be, it will require a lot of little models created to satisfy the diversity of the player base.
This is a much bigger suggestion than you realize. They should just add a loadout where you can pick a face from pre-created selection and bam, you have that face. I would rather have their time spent optimizing their technology and building features that add substance to gameplay.
EDIT: Grammar! Rawr!<!--QuoteEnd--></div><!--QuoteEEnd-->
whoa, didn't notice this one, hehe. *cracks knuckles*
*ahem* importing models: nah
why would we import models if you can just get the next best thing by importing the texture on top of a stock face model. There would be no model change at all, simply a re-texture. If all marine models have an option to have their face texture over-layed, then this would have the memory requirements of simply importing a texture from a connecting client. This would be simple to implement, and function just a simply.
Now you might be thinking, AWWWWWWW no variant in face models? then wont my textures seem out of place on top of the model? short answer: no. There is probably already a plan to add variants to the stock marine face models. Before overlaying your imported texture, the player could decide which model to overlay their texture onto. Of course this wouldnt be as unlimited as your hypothetical in my quote, but this would be the next best thing. I dont think anyone would complain if there were limited face models to lay their textures on because JUST the FEATURE ALONE of importing your face texture so others can see would be immensely satisfying to all parties in the server.
Another concern might be, what if the stock face model doesn't match up with the face texture, well this could simply helped by having an example face texture showing where all the parts of the face should be, to help people align their face texture to the face model.
I dont know why you wouldn't support this amazing idea, and if any devs are reading this, please, PLEASE consider this. I KNOW players would REALLY enjoy this feature. As I explained many times, this would not be very hard to implement.
edit: and when I mean stock facial variants, I dont mean replaceable parts of the face like the chin/ear/whatever, I mean entirely different and separate pre-made face models, all of which with varying facial features. We dont need interchangeable parts that bad.
Its about as complicated as custom sprays in half-life. UWE's engine is reported to have revolutionary infrastructure to allow on-the-fly downloading of game code to clients. With this already in place it should be trivial to implement.
If quality control is really an issue they could simply allow players to print logo's on the back of marine armor, or something "clean" along those lines.
Its about as complicated as custom sprays in half-life. UWE's engine is reported to have revolutionary infrastructure to allow on-the-fly downloading of game code to clients. With this already in place it should be trivial to implement.
If quality control is really an issue they could simply allow players to print logo's on the back of marine armor, or something "clean" along those lines.<!--QuoteEnd--></div><!--QuoteEEnd-->
Quality Control is a huge issue for me. I would like the immersion of having the game's graphics be consistent and there not be a huge drop in quality when I look at a marine's face. Would it be a nice feature down the road? Sure, but I do consider it a huge liability unless done with a amount of care that I don't think UWE has resources to do.
If the issue is that clone marines are an issue, than ideally there should be 16 variations of marine and players get assigned one randomly for the entire match. This lets them build a presence in the game for the duration of that round, but it does have the problem that 16 variations of marine would need to be created (e.g. white male 1, white male 2, black male 1, black male 2, white female 1, white female 2...etc) and that costs money.
@N3MES1S: Just placing a texture over a pre-existing model would just in result in incredibly ugly marines.
Having them be failugly is kinda the point anyways. Its more about trolling other people than anything. Half-life sprays never hurt anyone. :P
If the issue is that clone marines are an issue, than ideally there should be 16 variations of marine and players get assigned one randomly for the entire match. This lets them build a presence in the game for the duration of that round, but it does have the problem that 16 variations of marine would need to be created (e.g. white male 1, white male 2, black male 1, black male 2, white female 1, white female 2...etc) and that costs money.
@N3MES1S: Just placing a texture over a pre-existing model would just in result in incredibly ugly marines.<!--QuoteEnd--></div><!--QuoteEEnd-->
Yes, everyone has different opinions. This is why there are custom skin AND pure servers in team fortress 2/counter strike.
Obviously you would prefer a pure server to pertain your romantic attachment to being immersed into the game. I however will be playing this game for the lolz.
I wouldn't mind having the ability to pick from premade faces.