Looks like a good start. Some of your contours are a little shaky and the wings of the guard are a bit wonky. Its a good enough mesh to learn normal map baking though. A lot of the detail is hard to see in the pictures because there's no AO. What package did the images come from?
Ok. One thing you can do to improve your images is do a screen grab with nitrous/realistic turned on. That'll bring out more of the detail. Or you could just do a light tracer render. It's not really necessary this time, but it should help with future clay renders that you want crits on.
Very nice start. I really am a huge fan of Zelda (if you can't tell by my posts being in most Zelda threads ha ha), but one thing you might do is reconsider the hilt. You've got kind of a strange design. The rest of the model follows a pretty traditional Master Sword shape, but the "wings" don't. Here's what I'm talking about:
Something to consider if you are going for a fairly accurate take on the sword. Almost all the games from A Link to the Past to Skyward Sword have followed the same (or similar) design:
hey, I would accentuate the deepth of the detail, its really spot on to the reference, but to have it read from a distance, you'll need to accentuate this sort of thing. Otherwise spectacular modeling
I'd go in and smooth out that weird kink in the "wings" of the hilt. They're smooth on all the master sword ref that Bobthegreat sent you. But on yours they hit an obvious angle about half way down the "wing"
And like Rich Kid says, if most of the hilt has been modeled as one piece, that's a bit strange. You're more likely to get better AO and bakes, not to mention a much easier time modeling, if you model major shapes separately and then float them against each other.
my next step is to start cleaning up the wings i spent some time working on the low poly and in a couple days i cam going to fix the wings and clean it up a bit.
The wings on the high poly still aren't as smooth as they should be. The guard points much too high on the handle and should go out sideways more. I can understand the grip is more custom to you but as for as I see the pommel is nothing like any of the above concepts.
Even if you did model off that sword, the model is still not accurate to even that picture. The picture demonstrates exactly what I showed you in the pictures above. Keep working at it.
Mirror and overlap the islands that are the same. Unless you are going to make a part unique by adding a scratch or something, but even then. Mirror and overlap, it makes things much easier.
what particularly is off on the wings im not sure what your trying to show in the picture
You're using far too harsh angles. Both the wings and the parts that hold them are round, yours aren't. Also the wings get wider by the time they end, not skinnier like your model. Also the wings at the end may look like they wave in and out, it's just the angle of your reference. In fact they are flat.
You can see here where someone else did the same model with I'm assuming the same reference, note how his wings are round, and so are the sections holding it:
This follows much closer to the actual design of the sword through the years.
Ok so earlier in the thread you said you were mapping things uniquely to get some unique details in it... Where are they? Nothing stands out to me as being noticeable uniquely detailed. In which case you ought to have unwrapped only one side giving you a whole lot more texture rez to work with.
I can see two normal maps, an AO map, a diffuse map, a Normalmap that was run through the high pass filter or crazybump... I don't see a specular on that image...
The maps you used to create your final textures (AO, High-passed Normal etc) aren't needed at all on a final presentation image. The important part is the maps that are applied to the object in UDK (or Marmoset, whatever engine you're using), and even then some of the little shader specific stuff is often irrelevant. The important part is that you have on there your Normal, your Diffuse and your Specularity, the big three of current gen gaming.
Also when creating an image like that you want to include relevant data, Polycount Texture sizes etc. Because that helps us understand the level of quality too. If you can make something look at the same level of quality with half the polys and texture size that shows a higher level of skill etc.
All in all it's definitely a good first pass, take it as a learning experience, maybe make a couple edits (the grip could have a massive reduction in polys, you can use at least half that and achieve nearly the same silhouette). And then move on to something new and apply everything you've learned.
Your goal should be to turn this piece into the worst piece in your portfolio as fast as possible, by raising the quality of your new work.
Replies
The textures are going to make this amazing....I'm gonna check back on this one!
Something to consider if you are going for a fairly accurate take on the sword. Almost all the games from A Link to the Past to Skyward Sword have followed the same (or similar) design:
Everything else looks pretty good.
And like Rich Kid says, if most of the hilt has been modeled as one piece, that's a bit strange. You're more likely to get better AO and bakes, not to mention a much easier time modeling, if you model major shapes separately and then float them against each other.
The wings are still very blocky, which is the biggest flaw of the entire thing.
Keep Working on the high poly before anything.
Also it's Baked* not backed
Keep going!
You're using far too harsh angles. Both the wings and the parts that hold them are round, yours aren't. Also the wings get wider by the time they end, not skinnier like your model. Also the wings at the end may look like they wave in and out, it's just the angle of your reference. In fact they are flat.
You can see here where someone else did the same model with I'm assuming the same reference, note how his wings are round, and so are the sections holding it:
This follows much closer to the actual design of the sword through the years.
I can see two normal maps, an AO map, a diffuse map, a Normalmap that was run through the high pass filter or crazybump... I don't see a specular on that image...
The maps you used to create your final textures (AO, High-passed Normal etc) aren't needed at all on a final presentation image. The important part is the maps that are applied to the object in UDK (or Marmoset, whatever engine you're using), and even then some of the little shader specific stuff is often irrelevant. The important part is that you have on there your Normal, your Diffuse and your Specularity, the big three of current gen gaming.
Also when creating an image like that you want to include relevant data, Polycount Texture sizes etc. Because that helps us understand the level of quality too. If you can make something look at the same level of quality with half the polys and texture size that shows a higher level of skill etc.
All in all it's definitely a good first pass, take it as a learning experience, maybe make a couple edits (the grip could have a massive reduction in polys, you can use at least half that and achieve nearly the same silhouette). And then move on to something new and apply everything you've learned.
Your goal should be to turn this piece into the worst piece in your portfolio as fast as possible, by raising the quality of your new work.