Was there ever a Victoria 4.2 Lo Res Figure.There was for nearly all the other Daz Generation 1, 2 and 3 Figures! Cheers

Nope, never released.

I believe there were lod figures of v4 and m4.

Yes there are. 17k, 4k, 2k, and 1k LOD's. Never used them in Poser so not sure how to apply them in poser. They work in DAZ and Carrara natively. V4, M4, and K4 I think.

I think there is also a V4 Quick Start version which was based on the 17k version

Someone rerigged them, the full nick I don't remember but it starts with Joe.

You can use the reduce polygons function. It works reasonably well but AFAIK it reduces only the polygons, not the vertices.

FVerbaas posted at 7:00AM Tue, 08 September 2020 - #4398912

You can use the reduce polygons function. It works reasonably well but AFAIK it reduces only the polygons, not the vertices.

Isn't that in the gamedev version?

FVerbaas posted at 8:50AM Tue, 08 September 2020 - #4398912

You can use the reduce polygons function. It works reasonably well but AFAIK it reduces only the polygons, not the vertices.

Yup did that yesterday to V4, and it works ,didnt check verts or actual polys after , just wanted it reduced twice one @ the default amount and one a lot less (just to see how well it retained the morphs details I dialed in)

@ Liquid_Ice its in Poser Pro 11

Liquid_Ice posted at 3:56PM Tue, 08 September 2020 - #4398914

FVerbaas posted at 7:00AM Tue, 08 September 2020 - #4398912

Isn't that in the gamedev version?

It is in the current Poser 11 Pro.

Ah, good to know. Never needed though

No such thing as Poser Pro 11 anymore. It is officially Poser 11.

hborre posted at 11:00AM Tue, 08 September 2020 - #4398927

No such thing as Poser Pro 11 anymore. It is officially Poser 11.

What's in a name

Confusion.

CHK2033 posted at 11:37AM Tue, 08 September 2020 - #4398937

hborre posted at 11:28AM Tue, 08 September 2020 - #4398934

Confusion.

lol, true. although it says Poser Pro;

I am a little confused. Is the only version of Victoria 4 Los Res available in DAZ Studio only? Not for Poser! Cheers

here arrow

I dont use them so no idea how you would get them on her inside of Poser , but this is where there at.

The low res V4 came about due to the free content that came with DS2, DS1 had V3 RR and a fairy woodland scene, DS2 was a 17k poly V4 with a beach scene rather than the full mesh as an LE, the content we got with DS3 was pretty much the same as DS2.

If you have the free content for either DS2 or 3 then you have a CR2 that uses the 17K poly mesh, it'll take V4's textures, but not any of her morphs.

The LoD system in DS3 and 4 can transfer the active morphs to the LoD version, no idea if you can get them out of DS and into the Quick Start CR2 through.

You can always make morphs yourself. And btw they would also make great background or test models for animation.

And if you texture in a 3d painting app the los figure works much faster.

If you have that geometry and Poser 11 (Pro) you can build your lo-res version in the fitting room or the setup room.

You can simply edit a V4 cr2 so it points to one of the V4 LOD objects.

The original morphs of course will be useless then, but most can be re-transfered back inside Poser. (Provided you have PP 2014 or Poser 11)

As I disliked V4's original rigging, I deleted all the magnets and JCM's, and simply weightmapped my V4 LOD's from scratch.

Here's my weightmapped V4 17k LOD:

And here are some full bodymorphs I transfered over from regular V4:

Finally a custom morph sculpted after the original Stephanie Petite figure:

The process is pretty much identical to what I did with the M4 LOD meshes here:

https://www.renderosity.com/mod/forumpro/thread/print/?thread_id=2874173

These LOD's are of course easier to weightmap than the higher resolution figures, and thanks to the sub-D Morphbrush functionality included in Poser 11, can be just as detailed as the original ones.

Just for the record:

Here's V4 LOD 1K: Heart of gold, but a bit rough around the corners. Better render here with polygon smoothing rather than subdivision, as the SubD algorithm will make her look too skinny:

Here's V4 LOD 4K: Great personality, as she already has enough polygons in her head to have actual expression morphs. She has prop teeth and, like 1K above, no (sculpted) toes.

17K otoh is already "fully formed" and fully morphable. In fact, regular V4 is just a subdivided 17k with the subdivision permanently "baked" in.

Many thanks everyone. I found the OBJ's 1,2,4, and 17k.One last question if I may.I am a total dummy when it comes to fiddling with geometries.How do I weightmap these figures? Cheers

JP!!

Long time no see!

arrow1 posted at 9:25PM Thu, 10 September 2020 - #4399137

Many thanks everyone. I found the OBJ's 1,2,4, and 17k.One last question if I may.I am a total dummy when it comes to fiddling with geometries.How do I weightmap these figures? Cheers

Lol. Download Sasha16. No, Seriously.

Otherwise? I guess transfer V4's skeleton to the mesh in the setup or fitting room, or as Joe Public says have the cr2 reference the LOD mesh, delete all the magnets, merge zones to weightmaps, adjust her bends with the joint editor, create JCMs...

This might be of interest for starters...

https://www.renderosity.com/mod/forumpro/?thread_id=2868504&page_number=3

I rigged the 4k and 17k LOD versions of the V4 OBJ by loading the OBJ into a new Poser scene, switching to the Setup tab, and selecting the regular base V4 from the Figures library as the "donor" skeleton. Then I applied the Base and Morphs++ morph packages as per usual. Some of them continued to work, while others didn't, mainly in areas where the morph relied on more geometry than was present in the LOD version.

Aside from a weird "sticky vertex" issue in the lip that kicks in with a particular open mouth dial (can't recall which ATM), I was pleasantly surprised with how easy it was and how well it worked! Conforming clothes work just fine, as would dynamic, and of course hair always works (esp. if you always parent it rather than conforming rigged hair).

I don't recall about the 17k version, but I think the 4k version has fewer material zones, so even though existing skin materials will work, you might get a message about the material file having "missing zones." Because I was setting up these LOD versions for use as background characters, I just set my own lower-res skin materials and saved them specifically for use on the LOD versions.

I think that that the reason V4 came with OBJs in varying levels of detail is b/c DAZ Studio had a way to enable you to switch between the options, depending on how far in the background the figure would be. Poser doesn't have anything like that, but it's not that hard to just rig an LOD yourself, since you don't actually have to do any rigging; you're just copying the V4 rig onto a new OBJ.

Note: I think I tried using Sasha-16 as the donor at some point, in hopes that the LOD version would receive her weight-mapping, but it didn't work (nor should I have expected it to, really, but I learn a lot just by experimenting!)

Speaking of experimenting: if your goal is to have more V4's in a scene without having Poser slow down, then poly count probably isn't the problem. Based on my experiments, the slow down is far more likely to come from having lots of morph packages installed in each V4 AND having lots of conforming clothing items on each of them. Poser has to constantly think about what to do when any one of those morph dials is activated, and after a while it will have a brain meltdown! So the secret there is to keep each V4 to the bare minimum required to get the look you want.

Hello WandW! :-)

While I weightmapped my LOD's ffrom scratch "just for the heck of it", I think it's easiest to start with an already weightmapped V4. Weightmapps easily transfer from a higher res to a lower res figure of similar topology.

Sometimes there might be minimal editing needed, but nothing the morphbrush couldn't handle.

Same with morphs. For example the sticky vertice that perpetualrevision encountered while transferring over a smile morph, can easily be "tamed" with a little "smoothing" by the morphbrush.

In my 20yrs of Posering on low-end hardware, I think the biggest "resource hogs" are high res textures.Think about it: Unless you render your figure at super-high-end resolution at an insanely huge size, 90% of those 4K x 4K textures are just wasted, because the render engine just blurrs it all out.

So while I keep a copy of the original textures, of course, my "working textures" are 1200 x 1200 at most. Sometimes even less. (Unless it is a ground texture that can't be tiled)

Polygon reduction is still quite effective to save precious RAM, as Poser (Unlike Studio), loads all morphs it can find in a cr2 permanently into memory. So a high res figure with lots of morphs can quickly eat up a lot of memory before you even start rendering it.

JoePublic posted at 2:51AM Sun, 13 September 2020 - #4399196

In my 20yrs of Posering on low-end hardware, I think the biggest "resource hogs" are high res textures.Think about it: Unless you render your figure at super-high-end resolution at an insanely huge size, 90% of those 4K x 4K textures are just wasted, because the render engine just blurrs it all out.

So while I keep a copy of the original textures, of course, my "working textures" are 1200 x 1200 at most. Sometimes even less.

I've spent a lot of time trying to make Poser faster, and I found that, at least for my setup, textures were the second most likely cause for slowdowns (with lots of morphs + conforming apparel being the first). So I definitely agree with you there. I've gone through quite a few material setups and removed spec, bump, displacement, and/or normal maps to cut down on the amount of memory used by those images, replacing them with procedural alternatives. But I aim for a slightly stylized look in my renders, not for photorealism, and the procedural alternatives work well for that. They may not work as well, particularly for skin, for those whose goal is photorealistic close-up portraits.

Where I disagree is that 4k diffuse textures are wasted. For one thing, Firefly won't blur them out if (a) you turn off image filtering on all image maps (except for hair, which looks better with Quality filtering); and (b) you set the Post Filter Type to sinc and Post Filter Size to 2 or 3 in your Firefly render settings. Also, with computer monitors getting increasingly high-res, final renders should be larger (~3k or higher) or else they'll seem tiny and blurry in just a few years. I regularly do final renders at 4400 × 2475 pixels b/c I like to be able to see all the details and b/c it no longer takes ages to render that large. (My average for final renders is around 15 minutes for direct light.)

I actually wish more prop sets had higher res diffuse textures, so that they'd be useable for the foreseeable future without my needing to create new textures for them. I have quite a few older sets that I can no longer use in a "regular" render (i.e., one I don't intend to apply an art effect to) b/c the textures just aren't high enough quality, particularly compared to the textures on characters' skin and apparel. That doesn't apply to prop sets that make use of tiling textures, as those can easily be updated, but I have some that make use of painted-on detailing or other custom elements that I can't easily re-make in a higher res version. I've even tried out some apps that specialize in up-sizing photos w/o minimal loss of quality specifically so that I could up-size some of those textures!!

But I have also, in a few cases, made lower-res versions of textures just as you describe (while keeping the originals), and that's also a good thing to know how to do :-)

For textures resolution think of how many pixels the head of your figure will be in your final render, and how much you will see of any body or limbs texture in a figure that is fully dressed. I do not mention eyes. I do not mention specular, bump etc. maps.

There is a lot that can be done to free up space and speed up.

Makes me think of a script that does just that: swap textures to a size that makes sense for the project at hand. Problem is that Python access to the texturemaps cache is limited to zero.

PerpetualRevision said

"Where I disagree is that 4k diffuse textures are wasted. For one thing, Firefly won't blur them out if (a) you turn off image filtering on all image maps (except for hair, which looks better with Quality filtering); and (b) you set the Post Filter Type to sinc and Post Filter Size to 2 or 3 in your Firefly render settings. Also, with computer monitors getting increasingly high-res, final renders should be larger (~3k or higher) or else they'll seem tiny and blurry in just a few years. I regularly do final renders at 4400 × 2475 pixels b/c I like to be able to see all the details and b/c it no longer takes ages to render that large. (My average for final renders is around 15 minutes for direct light.)"

The way I look at it is, it's better to have high res textures and resize them down if need be, than have textures that need to be sized up. If someone is incapable of opening an image editor and changing an image size? If it's too much of an inconvenience? Imagine how inconvenient it would be if you had to create your textures and models from scratch? I know, a shocking proposition...

Actually, I'd say that crisp or none is best. Repeating or tiling textures that tend to have a Moiré pattern effect, higher res fabrics for instance, are best served by the slight anti aliasing that Quality texture filtering setting provides. For GI/INDIRECT lighting, seems like no texture filtering is best with hair. Or so I've read recommended in more than one instance. The Crisp setting was probably introduced when it was found that the anti aliasing happening with the existing texture filtering choices of fast and quality were causing seams to appear at uv margins on many textures. Mostly on those that had an insufficient bleed zone of pixels (which I recommend is between 6 and 10 pixels) at the boundaries of each shell or island.

Shader, bump, and displacement detail is better served by a lower min shading rate. I use 0.10 in render settings (default is 0.20) and batch modify every actor's rate via a script (individual min shading rate in properties) to match. Simply changing the global rate in render settings isn't enough, each actor must have a corresponding value setting if expected to be handled at that rate. In this way it's possible to have lower shading rates on things that would require it, and higher on less important things.

I think super high poly counts on models that work just as well with half the polys is some thing I still see a lot. Especially on toon figures that aren't going to be morphed as much as a "realistic" human figure that's expected to be many things to many people. And we have Subdivision and HD morphs now, but there's a dual edged sword factor with this (at this time) but it's definitely a step in the right direction.

On the topic of poly counts and Poser users I'd like to show you a highly rated environment uploaded to the Poser section of freebies at ShareCG, in fact the most highly rated Poser freebie in months... just download the model and load it via Wavefront obj import AS IS. It isn't in pp2 or cr2 format. In fact it's been exported out of DS (as indicated by comment lines in the obj) and put in a folder and plopped in the Poser section. Apparently it loads fine in DS. Not so much anywhere else (except maybe Vue). Unfortunately it wasn't uploaded to the DS format section, only the Poser section. Tell me if there's anything unusual about this very interesting model, easily one of the weirdest things I've encountered. Oh btw, It's a couple of million polys or more, I stopped trying to figure out how many exactly after I realized that a couple of included open source 3d scan decorations hidden in there were breaking the 2 million mark. This is what Poser users want if popularity is any indication...

https://sharecg.com/v/96213/browse/11/Poser/Temple-of-Venus

On the topic of poly counts and Poser users I'd like to show you a highly rated environment uploaded to the Poser section of freebies at ShareCG, in fact the most highly rated Poser freebie in months... just download the model and load it via Wavefront obj import AS IS. It isn't in pp2 or cr2 format. In fact it's been exported out of DS (as indicated by comment lines in the obj) and put in a folder and plopped in the Poser section. Apparently it loads fine in DS. Not so much anywhere else (except maybe Vue). Unfortunately it wasn't uploaded to the DS format section, only the Poser section. Tell me if there's anything unusual about this very interesting model, easily one of the weirdest things I've encountered. Oh btw, It's a couple of million polys or more, I stopped trying to figure out how many exactly after I realized that a couple of included open source 3d scan decorations hidden in there were breaking the 2 million mark. This is what Poser users want and can handle if popularity is any indication...

https://sharecg.com/v/96213/browse/11/Poser/Temple-of-Venus

EldritchCellar posted at 4:27PM Sun, 13 September 2020 - #4399395

On the topic of poly counts and Poser users I'd like to show you a highly rated environment uploaded to the Poser section of freebies at ShareCG, in fact the most highly rated Poser freebie in months... just download the model and load it via Wavefront obj import AS IS. It isn't in pp2 or cr2 format. In fact it's been exported out of DS (as indicated by comment lines in the obj) and put in a folder and plopped in the Poser section. Apparently it loads fine in DS. Not so much anywhere else (except maybe Vue). Unfortunately it wasn't uploaded to the DS format section, only the Poser section. Tell me if there's anything unusual about this very interesting model, easily one of the weirdest things I've encountered. Oh btw, It's a couple of million polys or more, I stopped trying to figure out how many exactly after I realized that a couple of included open source 3d scan decorations hidden in there were breaking the 2 million mark. This is what Poser users want and can handle if popularity is any indication...

https://sharecg.com/v/96213/browse/11/Poser/Temple-of-Venus

I just tried it in Poser, and it actually works pretty well. Renders very quickly, even in Superfly, perhaps because the textures are simple.

Personally, I prefer this kind of item to be in separate pieces, so you can only load the parts you need, but this is really a pretty amazing item for free. That's probably why people are rating it so highly: they recognize the amount of work that went into it.

Yup. Nevermind.

Mostly clothed La Femme in the Temple of Venus with a sword...

I wonder what he's using for modeling this stuff. He's churning them out at a pretty good clip. The only software he lists in his profile are DAZ Studio, Poser, Photoshop, and Bryce. He says he had to export the OBJ of his factory model to Bryce because it was too big for whatever software he was using. That's why the scaling is off. (Though you'd think that would be an easy thing to fix.)

"You can use the reduce polygons function."

Huh ? Where did that spring from ??

After picking jaw off floor, I went hunting through PDF manual, found documented on page 999~~1000 etc. If I can persuade this to work, I may be able to salvage some too-many-poly terrain mega-meshes...

Thank you, FVerbaas !!