Aller au contenu

Photo

Fan-Made Fix For Face Modification & Import Bugs Remaining After Patch 1.02


  • Ce sujet est fermé Ce sujet est fermé
2037 réponses à ce sujet

#826
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages
Yes, I have seen that. :)  I came to that conclusion independently.

Modifié par Vhalkyrie, 14 avril 2012 - 04:02 .


#827
StarAce905-2

StarAce905-2
  • Members
  • 321 messages

Vhalkyrie wrote...

I found this and it confirms that I thought. They altered the geometry in such a way that slider positions will give off results.

Example:

Mass Effect 1's "Mouth Shape 5" gives the top lip a value of <6, 4, 160>.
Mass Effect 2's "Mouth Shape 5" gives the top lip a value of <6.5, 4.1, 161>.
Mass Effect 3's "Mouth Shape 5" gives the top lip a value of <6.3, 4.2, 160.75>.

When the face importer reads in value <6.5, 4.1, 161>, it uses the slider position equivalent in ME3 <6.3, 4.2, 160.75> by rounding to closest.  For some facial features, this could occur by a larger margin.  For some faces that are close, adjusting by a tick in either direction could make a closer match.  In others, it won't make a difference because they widened/shortened the base feature by a noticable degree.

http://faceeffect.bl...you-put-it.html

ME2 gave us exact matches because they imported the raw data, which was why the "Accepted face" wasn't editable.  If you wanted to change it, you had to go through custom face again and rebuild.  In ME3, it doesn't import raw data.  It uses the nearest slider positions.

This is not quite the problem. ME3 also imports raw data properly. Open up an ME2 headmorph on ME2 Gibbed, and open its import into ME3 with ME3 Gibbed and also check the headmorph. The values are there. (Though I'm unsure whether all are. Possibly, one or two variables are missing in ME3, but that could be an oversight on my part. Have to check again.) But it seems ME3 interprets these values in a different manner.

#828
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages

StarAce905-2 wrote...
This is not quite the problem. ME3 also imports raw data properly. Open up an ME2 headmorph on ME2 Gibbed, and open its import into ME3 with ME3 Gibbed and also check the headmorph. The values are there. (Though I'm unsure whether all are. Possibly, one or two variables are missing in ME3, but that could be an oversight on my part. Have to check again.) But it seems ME3 interprets these values in a different manner.


The values in my ME2 versus ME3 headmorphs were different by a statistical margin.  The values are there, but they are not the same.  It doesn't import the ME2 raw data, it uses the ME3 slider positions, and stores those.  For many features, I am able to copy the ME2 data to ME3 and get a closer result.

Modifié par Vhalkyrie, 14 avril 2012 - 04:26 .


#829
Mingolo

Mingolo
  • Members
  • 113 messages

StarAce905-2 wrote...

Vhalkyrie wrote...

I found this and it confirms that I thought. They altered the geometry in such a way that slider positions will give off results.

Example:

Mass Effect 1's "Mouth Shape 5" gives the top lip a value of <6, 4, 160>.
Mass Effect 2's "Mouth Shape 5" gives the top lip a value of <6.5, 4.1, 161>.
Mass Effect 3's "Mouth Shape 5" gives the top lip a value of <6.3, 4.2, 160.75>.

When the face importer reads in value <6.5, 4.1, 161>, it uses the slider position equivalent in ME3 <6.3, 4.2, 160.75> by rounding to closest.  For some facial features, this could occur by a larger margin.  For some faces that are close, adjusting by a tick in either direction could make a closer match.  In others, it won't make a difference because they widened/shortened the base feature by a noticable degree.

http://faceeffect.bl...you-put-it.html

ME2 gave us exact matches because they imported the raw data, which was why the "Accepted face" wasn't editable.  If you wanted to change it, you had to go through custom face again and rebuild.  In ME3, it doesn't import raw data.  It uses the nearest slider positions.

This is not quite the problem. ME3 also imports raw data properly. Open up an ME2 headmorph on ME2 Gibbed, and open its import into ME3 with ME3 Gibbed and also check the headmorph. The values are there. (Though I'm unsure whether all are. Possibly, one or two variables are missing in ME3, but that could be an oversight on my part. Have to check again.) But it seems ME3 interprets these values in a different manner.


No, the values are different, like Vhalkyrie points out. We actually went through all this a few weeks ago, and like I said, Phoenix and Malcroix posted how to change the raw variables using gibbed, because they are different in ME3. Even doing this though, doesn't always work as well as you'd expect, because LODs is what matter and they're harder to translate. Plus textures also making a big difference, and they're harder to change as well.

Modifié par Mingolo, 14 avril 2012 - 04:20 .


#830
StarAce905-2

StarAce905-2
  • Members
  • 321 messages

Vhalkyrie wrote...



The values in my ME2 versus ME3 headmorphs were different by a statistical margin.  The values are there, but they are not the same.  It doesn't import the ME2 raw data, it uses the ME3 slider positions, and stores those.  For many features, I am able to copy the ME2 data to ME3 and get a closer result.

It seems I only checked the MorphFeatures part. Sorry, ignore what I said, please. :)

#831
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages

Mingolo wrote...

No, the values are different, like Vhalkyrie points out. We actually went through all this a few weeks ago, and like I said, Phoenix and Malcroix posted how to change the raw variables using gibbed, because they are different in ME3. Even doing this though, doesn't always work as well as you'd expect, because LODs is what matter and they're harder to translate. Plus textures also making a big difference, and they're harder to change as well.


Do you have links to these?  Because it would be good to leverage data that already exists.

[Edit] If you're referring to Face Effect, already got it.

Modifié par Vhalkyrie, 14 avril 2012 - 04:26 .


#832
Fidget6

Fidget6
  • Members
  • 2 437 messages
Some of them look different, but others look almost identical. Some of them look like they've aged a bit, but it's been several years between ME1 and ME3 so.....

#833
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages

StarAce905-2 wrote...

Vhalkyrie wrote...



The values in my ME2 versus ME3 headmorphs were different by a statistical margin.  The values are there, but they are not the same.  It doesn't import the ME2 raw data, it uses the ME3 slider positions, and stores those.  For many features, I am able to copy the ME2 data to ME3 and get a closer result.

It seems I only checked the MorphFeatures part. Sorry, ignore what I said, please. :)


MorphFeatures doesn't appear to do anything, and I've seen multiple sources that corroborate that.  Which is too bad because it seems like it had a lot of fine control possibilities.

#834
Mingolo

Mingolo
  • Members
  • 113 messages

Vhalkyrie wrote...

Do you have links to these?  Because it would be good to leverage data that already exists.

[Edit] If you're referring to Face Effect, already got it.


The link I posted to Phoenix's tutorial has some, but most of the discussions as Phoenix and other people tested stuff went on in the locked thread, buried somewhere in the low 100s pages. Would be near impossible to find now.

Modifié par Mingolo, 14 avril 2012 - 04:33 .


#835
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages
Well here is what I've found so far. LOD cannot be translated directly but this has already been established. Choosing overall base structure in the CC gives a better starting point for LOD match. This may mean deviating from the slider positions, even if the face code is a match. This can be fine controlled with by changing the OffsetBones, which also may need adjusting. OffsetBones seems to be the variable positions that are controlled by the sliders, allowing customization. OffsetBones parameters will give different results for different base head meshes, because each base head mesh has different starting coordinates. It is what it says, which is an "offset". Some start off with higher cheekbones than others, for example.

If this has already been discovered, then consider it independent verification.  I hadn't seen enough that was addressing my specific issue, so I dug into it myself.

Modifié par Vhalkyrie, 14 avril 2012 - 04:41 .


#836
Mingolo

Mingolo
  • Members
  • 113 messages
Ahh, so if the starting head mesh is different in the ME3 CC in the first place, then you'll still not get the same face. That may be why it doesn't work for some. I really got lost in Phoenix's complicated explanation of why it didn't work as well always, lol.

Anyway, Woodbyte and his computer geek friends were working on mapping the ME2 and ME3 LODs in a 3D program, to see if they could be translated and directly imported. Last we heard anyway. That's my only hope right now.

Modifié par Mingolo, 14 avril 2012 - 04:53 .


#837
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages
Yeah, I think the LOD mapping is way more detail then I want to get into - lol. I am eager for their results.

Well I can't say for sure yet, but I am theorizing that is why translating the offset bones from ME2 to ME3 doesn't give good results for some people like Clarian. The 'offset' isn't at a relative starting point to the original.

#838
ThinkIntegral

ThinkIntegral
  • Members
  • 471 messages
Vhalkyrie have you spoken with woodbyte about what he's doing? He's attempting to map the LODs in ME3, not sure if you saw his post back in the other thread...

#839
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages

ThinkIntegral wrote...

Vhalkyrie have you spoken with woodbyte about what he's doing? He's attempting to map the LODs in ME3, not sure if you saw his post back in the other thread...


I saw something about it, but haven't followed it closely.

#840
LilyasAvalon

LilyasAvalon
  • Members
  • 5 076 messages
There's something sad how much Shepard changes from ME2 to ME3.

#841
ThinkIntegral

ThinkIntegral
  • Members
  • 471 messages

Vhalkyrie wrote...

ThinkIntegral wrote...

Vhalkyrie have you spoken with woodbyte about what he's doing? He's attempting to map the LODs in ME3, not sure if you saw his post back in the other thread...


I saw something about it, but haven't followed it closely.


social.bioware.com/social.bioware.com/forum/1/topic/323/index/9661093/265#10835955

Later he said he sort of has figured out how the mapping differs between the two games and that ME3 gets some stuff right. Since I don't know graphics I'm not sure what he can do for certain. Maybe you could speak with him?

#842
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages

ThinkIntegral wrote...

social.bioware.com/social.bioware.com/forum/1/topic/323/index/9661093/265#10835955

Later he said he sort of has figured out how the mapping differs between the two games and that ME3 gets some stuff right. Since I don't know graphics I'm not sure what he can do for certain. Maybe you could speak with him?


Looks interesting. He probably already knows what I do, but wouldn't hurt to see if there something I can offer.  Thanks for the tip.

#843
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages
Here is my current progress.  Some detail is lost due to the scaling, but hope you can get the idea.  It actually is a lot closer than this in the larger view.  The eyes on the right look a bit different because she is squinting at Traynor and/or caught in mid blink.  The differences in game are narrowing.

ME2 -> ME3 Base Import -> ME3 Remapped

Posted Image

Larger view:

Posted Image

Unfortunately I don't have this process down in a form where I can easily tell you how to do it for your own games. If I can do this by hand, then it is certainly possible for Bioware to fix. I have some real life projects I have to get done, so hopefully I'll get back to this soon.  There is hope!

Does anybody have the Gibbed codes for the same color hair on the left?

Modifié par Vhalkyrie, 14 avril 2012 - 06:35 .


#844
DaylaLayne

DaylaLayne
  • Members
  • 82 messages
I find it strange that my Jane looks almost perfect on import but the other look kinda not right. I think the eyes are low and lips are high. I don't know drives me kinda crazy though.

#845
StarAce905-2

StarAce905-2
  • Members
  • 321 messages
Okay, so I followed what was in the Face Effect blog, and Shepard turned out even uglier. :(

#846
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages
I just read the entry on Face Effect about deleting LODs and relationship to MorphFeatures. This might be the ticket I need. Arg...got to get my other stuff done...

#847
Vhalkyrie

Vhalkyrie
  • Members
  • 1 917 messages

StarAce905-2 wrote...

Okay, so I followed what was in the Face Effect blog, and Shepard turned out even uglier. :(


It has better results if you have a fairly close match to begin with.  It is more fine tuning, rather than a fix.

#848
Mingolo

Mingolo
  • Members
  • 113 messages
I believe the deletion of the LODs also lead nowhere. Phoenix tried and it it apparently kind of worked for her character by pure luck, but others (including Clarian) got really off results that looked nothing like their characters (I think the consensus in the thread was that when you delete the LODs, the game recreates them, using a default head mesh). Tell us if you find anything new though.

Modifié par Mingolo, 14 avril 2012 - 07:06 .


#849
StarAce905-2

StarAce905-2
  • Members
  • 321 messages
Ah, alright. I'll give up for now and continue with the new femShep I have. By the way, is it just me, or is the new classy hair just... horrendous? It seems to have a weird ability to make most femSheps look ugly. -__-

#850
Terror_K

Terror_K
  • Members
  • 4 362 messages

Chris Priestly wrote...


Hi Everyone.
 
Mass Effect 3 Patch 2 is now available for the PC, Xbox 360 and PS3. You can see a full list of fixes here. One of the most important fixes was for people who could not import their Commander Shepard face codes from Mass Effect 2 into Mass Effect 3, this issue is resolved with Patch 2 and fans can now import their Shepard’s faces into Mass Effect 3.
 
I am seeing people saying that even though the faces can import, they do not look identical to how they appeared in Mass Effect 2. While the face codes do now correctly import into Mass Effect 3, due to improvements in things like shaders, lighting, textures, etc they will not look identical to the faces in Mass Effect 2. This was the same situation when players imported faces from Mass Effect 1 into Mass Effect 2.  The team has worked very hard to make sure that faces look as close to previous versions as possible. If there are small details you want to adjust, remember you can adjust your faces after importing before you start playing the game.
 
If your face has significant changes or you cannot import your save, please check these steps:

  • The saved game you are importing should be the final saved game from the end of Mass Effect 2.
  • The saved game should not have been modified in any way.
  • If you are using the correct saved game and have not modified your face code, please report your problem in the Technical Support forum for your platform or contact our Customer Service. Thanks.

:devil:


Wow... you guys really have justed checked out and don't care, huh? Bad enough that it wasn't going at launch, then we get a patch that doesn't even fix the issue, and fans discover that some faces just can't be recreated because whoever was in charge of the CC didn't make sure that every facial option was available, and now after all this you can't even admit to the screw up and just sweep it under the rug, feigning ignorance and pretending the issue doesn't exist and is something it's not.

Glad I'm leaving BioWare behind now with this quality of customer service and attitude. It's pretty clear this will never really be fixed.