Daz Studio Pro BETA - version 4.22.0.15! (*UPDATED*)

nonesuch00

PerttiA said:

nonesuch00 said:

I always install via DIM but for whatever reason sometimes DAZ Studio insists on creating that DAZ Connect content directory and sending some updates/contents for some packages that get duplicated in that connect directory. Are you saying that's the problem? How can I stop that? Delete the entire connect content directory in the File Manager? Or in the DAZ Connect pane?

Stop logging in to Daz Studio, the option to log in, is there to enable using Daz Connect for updating/installing products. 

DS Preferences -> Startup options, uncheck everything.

The official method of getting rid of Daz Connect highjacked installations, is to uninstall them in Smart Content.

If you have allowed Daz Connect to make updates/installations, those will get priority over DIM-installed items, even in cases when Cinnect installations do noe work.

Ah OK, thanks! I will be glad to see the back side of DAZ Connect!

Richard Haseltine

PerttiA said:

nonesuch00 said:

I always install via DIM but for whatever reason sometimes DAZ Studio insists on creating that DAZ Connect content directory and sending some updates/contents for some packages that get duplicated in that connect directory. Are you saying that's the problem? How can I stop that? Delete the entire connect content directory in the File Manager? Or in the DAZ Connect pane?

Stop logging in to Daz Studio, the option to log in, is there to enable using Daz Connect for updating/installing products. 

That isn't entirely true, and may become elss true as features are added. Note that DS no longer updates metadata by default, so the old issue with Connect metadata potentially replacing the current metadata has gone.

DS Preferences -> Startup options, uncheck everything.

The official method of getting rid of Daz Connect highjacked installations, is to uninstall them in Smart Content.

If you have allowed Daz Connect to make updates/installations, those will get priority over DIM-installed items, even in cases when Cinnect installations do noe work.

barbult

Thank you Saxa -- SD and crosswind for your insights on the NVIDIA driver issues. I went to the NVIDIA site to regress the driver and saw that there was yet another new studio driver version, 546.33. I decided to try it first, in case there was an obvious bug that they fixed. So far, so good. I haven't even been able to get DS into a Not Responding state yet with this driver, when it was happening very frequently with the 546.01 version. For now, I'll just keep an eye on it and see what happens. If the problem reappears, I'll regress the driver.

jbowler

Richard Haseltine said:

jbowler said:

Nested directory encountered while adding a content folder: G:/DAZStudio/Genesis-81 : G:/DAZStudio/Genesis-8

I have two CMS directories, one for explicit 8.1 content, calls ".../Genesis-81" and the other for G8 content ".../Genesis-8".  This would seem to be a problem, "WARNING", what are the rules?  Is it no-leading-prefix (so I have to rename ".../Genesis-8" as ".../Genesis-80"?)

Is G:/DAZStudio/ also a content directory? 

No.  All my CMS directories are at the same level; G:\DAZStudio\<dir> where <dir> is a different name for each CMS directory.  There is no "nesting" in the conventional computer sense and the only case where one CMS directory path (or name) is a substring of another is this one.

 

PerttiA

jbowler said:

Richard Haseltine said:

jbowler said:

Nested directory encountered while adding a content folder: G:/DAZStudio/Genesis-81 : G:/DAZStudio/Genesis-8

I have two CMS directories, one for explicit 8.1 content, calls ".../Genesis-81" and the other for G8 content ".../Genesis-8".  This would seem to be a problem, "WARNING", what are the rules?  Is it no-leading-prefix (so I have to rename ".../Genesis-8" as ".../Genesis-80"?)

Is G:/DAZStudio/ also a content directory? 

No.  All my CMS directories are at the same level; G:\DAZStudio\<dir> where <dir> is a different name for each CMS directory.  There is no "nesting" in the conventional computer sense and the only case where one CMS directory path (or name) is a substring of another is this one.

Show us the Content Directories in Content Directory Manager 

Artini

I can confirm that the latest Daz Studio beta (4.22.0.12) has fixed the issue introduced in 4.22.0.10

when Daz Studio beta has crashed when trying to save characters created with Face Transfer 2.

Thanks a lot for fixing the issue.

 

silverbishounen_86939832fd

I set up my firewall to block DS because I'm on a metered connection, but the (incredibly annoying) New Scene tab is doing an end run around the firewall, so if I want to use the current version of DS I have to either swap to the less secure free wifi I have access to, or risk going over my download limit.

I don't want to shop while I'm running DS. I'm in creative mode. I don't want to be advertised to, I'm in creative mode. I load the Daz Store regularly to look for new items that might interest me, check the lightning deals, etc. I don't need a non-functional store pop up ad that I can't even turn off.

jbowler

PerttiA said:

Show us the Content Directories in Content Directory Manager 

I find it completely trivial to get the error message; I just create a new directory, rename it so that the name (and the path name) is a suffx'ed version of an existing name (and path name) then boot Studio.  Here, using a different name and a fairly minimal config (just the base programs and 'starter essentials'), BaseFooBar is completely empty:

2023-12-14 16:12:57.412 +++++++++++++++ DAZ Studio 4.22.0.12 starting +++++++++++++++++
[ * * * ]
2023-12-14 16:12:57.665 [INFO] :: Scanning Content folders...
2023-12-14 16:12:57.665 [WARNING] :: \src\sdksource\general\dzcontentmgr.cpp(1851): Nested directory encountered while adding a content folder: G:/DAZStudio/BaseFooBar : G:/DAZStudio/Base
2023-12-14 16:12:57.717 [INFO] :: Creating Render Manager...

 

Here's the screenshot:

I don't understand why other people don't see this; the "WARNING" is pretty much at the start of the log file.  So there's some reason the CMS system thinks it can't handle two CMS directories one of which is a leading substring of the other.  Maybe it's using the full path name with path name separators stripped.  It doesn't matter much; I just want to know what the consequences are and what the actual rules are for CMS directories.  The fact that the error message is wrong and clearly so (the two directories are not "nested") makes me wonder if there is a bug in DS.

 

Richard Haseltine

jbowler said:

PerttiA said:

Show us the Content Directories in Content Directory Manager 

I find it completely trivial to get the error message; I just create a new directory, rename it so that the name (and the path name) is a suffx'ed version of an existing name (and path name) then boot Studio.  Here, using a different name and a fairly minimal config (just the base programs and 'starter essentials'), BaseFooBar is completely empty:

2023-12-14 16:12:57.412 +++++++++++++++ DAZ Studio 4.22.0.12 starting +++++++++++++++++
[ * * * ]
2023-12-14 16:12:57.665 [INFO] :: Scanning Content folders...
2023-12-14 16:12:57.665 [WARNING] :: \src\sdksource\general\dzcontentmgr.cpp(1851): Nested directory encountered while adding a content folder: G:/DAZStudio/BaseFooBar : G:/DAZStudio/Base
2023-12-14 16:12:57.717 [INFO] :: Creating Render Manager...

 

Here's the screenshot:

I don't understand why other people don't see this; the "WARNING" is pretty much at the start of the log file.  So there's some reason the CMS system thinks it can't handle two CMS directories one of which is a leading substring of the other.  Maybe it's using the full path name with path name separators stripped.  It doesn't matter much; I just want to know what the consequences are and what the actual rules are for CMS directories.  The fact that the error message is wrong and clearly so (the two directories are not "nested") makes me wonder if there is a bug in DS.

I see this too, please report it to Daz via a ticket.

Saxa -- SD

barbult said:

Thank you Saxa -- SD and crosswind for your insights on the NVIDIA driver issues. I went to the NVIDIA site to regress the driver and saw that there was yet another new studio driver version, 546.33. I decided to try it first, in case there was an obvious bug that they fixed. So far, so good. I haven't even been able to get DS into a Not Responding state yet with this driver, when it was happening very frequently with the 546.01 version. For now, I'll just keep an eye on it and see what happens. If the problem reappears, I'll regress the driver.

Just checked.  546.01 was the issue for me too.  Updated to that cos wanted to test the new Nvidia feature of SysMem.  Both the Nvidia game drivers newer then that were fine so far.

What have noticed with Sysmem on, is that before when DS could no longer launch Iray (stuck in neverending loading loop) - onlyfix was to terminate DS and start fresh.  Now can just toggle drawstyles-back and forth  and Iray render is back.  That is priceless in itself not having to reload large scenes.

crosswind

barbult said:

Thank you Saxa -- SD and crosswind for your insights on the NVIDIA driver issues. I went to the NVIDIA site to regress the driver and saw that there was yet another new studio driver version, 546.33. I decided to try it first, in case there was an obvious bug that they fixed. So far, so good. I haven't even been able to get DS into a Not Responding state yet with this driver, when it was happening very frequently with the 546.01 version. For now, I'll just keep an eye on it and see what happens. If the problem reappears, I'll regress the driver.

 NO proplem ~ Pls let us know if everything works fine with this version on you side.

Imago

silverbishounen_86939832fd said:

 I don't need a non-functional store pop up ad that I can't even turn off.

+1

We really need an option to turn off that invasive "new scene" window and revert to the old immediate one, along with the classic "recent" menu.

AliKat

Imago said:

silverbishounen_86939832fd said:

 I don't need a non-functional store pop up ad that I can't even turn off.

+1

We really need an option to turn off that invasive "new scene" window and revert to the old immediate one, along with the classic "recent" menu.''

+2

no__name

 

4.5 Ghost lights

Area light sources in Iray Photoreal are created via regular scene geometry. As such, these are typically always visible, both directly (from the camera) and indirectly, meaning in reflection or refraction, and do cast shadows. This may in some situations be undesirable, in particular if light sources are placed by artists to create a certain look or lighting mood: the intention may then be that lights should illuminate the scene (to a large extent), but also should otherwise not show up directly in the rendering. This creates a new type of problem for physically based rendering though: When is the contribution of a light source considered to be wanted, and when should it be omitted, without jeopardizing the rest of the light transport simulation?

To get started, the basic building blocks that control the visibility of lights are:

• For camera rays the visible attribute may be used to exclude directly visible geometry.
• A transparent MDL material (thin-walled, with only a specular transmission BSDF) can be used to make the non-emissive part of the light geometry disappear, including shadows.
• Light path expressions can be used to classify and (exclude) certain light contribution paths from the rendering.

Iray Photoreal further adds the concept of ghost lights to complement these basic concepts. Ghost lights allow for more fine-grained control on how light sources are visible in glossy interactions. Such lights feature an additional factor to enable a smooth blend between full and no contribution in glossy reflection and transmission. The factor can be set as an attribute on the IAttribute_set class and may affect both objects emitting and objects receiving light:

mi::Float32 ghostlight_factor = 1.0

Values greater than one turn an area source of light into a ghost light. This will immediatelly remove contributions of that light source from paths reaching the light via pure specular reflection and transmission. The larger the value, the more of the emission from that source will be removed from glossy interactions, i.e. glossy highlights and reflections from that source will smoothly be faded out. Increasing the factor by one will halve the peak brightness of glossy highlights. Thereby, surfaces with higher roughness are influenced less. In any case, diffuse materials are excluded from the factor and will always be lit from the light source the same way. Note that this factor can also be set on the receiving object side, where it acts as a multiplier to the light source factor (but only if there are ghost lights in the scene to begin with). This allows to fine-tune the behavior on a per object basis to achieve both a stronger removal of highlights or their recovery by using factors less than one.

 

I tested this out of curiosity (via freshly updated advance Iray node propriety), and it's very possible I miss something, but it still not working correctly.

I tried a bit everything under the sun, but pseudo 'ghost light' still cast shadows :

 

Richard Haseltine

AliKat said:

Imago said:

silverbishounen_86939832fd said:

 I don't need a non-functional store pop up ad that I can't even turn off.

+1

We really need an option to turn off that invasive "new scene" window and revert to the old immediate one, along with the classic "recent" menu.''

+2

If you scroll back you will see methods of doing so.

Richard Haseltine

no__name said:

 

4.5 Ghost lights

Area light sources in Iray Photoreal are created via regular scene geometry. As such, these are typically always visible, both directly (from the camera) and indirectly, meaning in reflection or refraction, and do cast shadows. This may in some situations be undesirable, in particular if light sources are placed by artists to create a certain look or lighting mood: the intention may then be that lights should illuminate the scene (to a large extent), but also should otherwise not show up directly in the rendering. This creates a new type of problem for physically based rendering though: When is the contribution of a light source considered to be wanted, and when should it be omitted, without jeopardizing the rest of the light transport simulation?

To get started, the basic building blocks that control the visibility of lights are:

• For camera rays the visible attribute may be used to exclude directly visible geometry.
• A transparent MDL material (thin-walled, with only a specular transmission BSDF) can be used to make the non-emissive part of the light geometry disappear, including shadows.
• Light path expressions can be used to classify and (exclude) certain light contribution paths from the rendering.

Iray Photoreal further adds the concept of ghost lights to complement these basic concepts. Ghost lights allow for more fine-grained control on how light sources are visible in glossy interactions. Such lights feature an additional factor to enable a smooth blend between full and no contribution in glossy reflection and transmission. The factor can be set as an attribute on the IAttribute_set class and may affect both objects emitting and objects receiving light:

mi::Float32 ghostlight_factor = 1.0

Values greater than one turn an area source of light into a ghost light. This will immediatelly remove contributions of that light source from paths reaching the light via pure specular reflection and transmission. The larger the value, the more of the emission from that source will be removed from glossy interactions, i.e. glossy highlights and reflections from that source will smoothly be faded out. Increasing the factor by one will halve the peak brightness of glossy highlights. Thereby, surfaces with higher roughness are influenced less. In any case, diffuse materials are excluded from the factor and will always be lit from the light source the same way. Note that this factor can also be set on the receiving object side, where it acts as a multiplier to the light source factor (but only if there are ghost lights in the scene to begin with). This allows to fine-tune the behavior on a per object basis to achieve both a stronger removal of highlights or their recovery by using factors less than one.

 

I tested this out of curiosity (via freshly updated advance Iray node propriety), and it's very possible I miss something, but it still not working correctly.

I tried a bit everything under the sun, but pseudo 'ghost light' still cast shadows :

That looks to me like a darker patch due to being edge on to the light source, not a cast shadow. If the source is planar then the area immediately below will not get any light  (because it would be hit only by rays coming out parallel to the plane) and as you move along the ground the light flux will increase - if the plane is emitting from front and back that will yield a dark band, as you have here. This is, if I am reading it correctly, the expected outcome of the geometry.

no__name

Richard Haseltine said:

That looks to me like a darker patch due to being edge on to the light source, not a cast shadow. If the source is planar then the area immediately below will not get any light  (because it would be hit only by rays coming out parallel to the plane) and as you move along the ground the light flux will increase - if the plane is emitting from front and back that will yield a dark band, as you have here. This is, if I am reading it correctly, the expected outcome of the geometry.

From my quick testing me it's clearly the mesh shadow, but again I may be wrong.
That said, either way, it completely defeats the purpose tho.

Can someone share a save with the same type of scene without the shadow with the new advanced Iray node ?

Richard Haseltine

no__name said:

Richard Haseltine said:

That looks to me like a darker patch due to being edge on to the light source, not a cast shadow. If the source is planar then the area immediately below will not get any light  (because it would be hit only by rays coming out parallel to the plane) and as you move along the ground the light flux will increase - if the plane is emitting from front and back that will yield a dark band, as you have here. This is, if I am reading it correctly, the expected outcome of the geometry.

From my quick testing me it's clearly the mesh shadow, but again I may be wrong.
That said, either way, it completely defeats the purpose tho.

Can someone share a save with the same type of scene without the shadow with the new advanced Iray node ?

I still don't think it is a shadow, it is a lack of light being cast in that direction. If you want all-round illumination you need a different configuration of sources - try a cube, with only one division that would be only six fces (and it could be single-sdied).

barbult

crosswind said:

barbult said:

Thank you Saxa -- SD and crosswind for your insights on the NVIDIA driver issues. I went to the NVIDIA site to regress the driver and saw that there was yet another new studio driver version, 546.33. I decided to try it first, in case there was an obvious bug that they fixed. So far, so good. I haven't even been able to get DS into a Not Responding state yet with this driver, when it was happening very frequently with the 546.01 version. For now, I'll just keep an eye on it and see what happens. If the problem reappears, I'll regress the driver.

 NO proplem ~ Pls let us know if everything works fine with this version on you side.

So far everything is fine with studio driver 546.33.

crosswind

barbult said:

crosswind said:

barbult said:

Thank you Saxa -- SD and crosswind for your insights on the NVIDIA driver issues. I went to the NVIDIA site to regress the driver and saw that there was yet another new studio driver version, 546.33. I decided to try it first, in case there was an obvious bug that they fixed. So far, so good. I haven't even been able to get DS into a Not Responding state yet with this driver, when it was happening very frequently with the 546.01 version. For now, I'll just keep an eye on it and see what happens. If the problem reappears, I'll regress the driver.

 NO proplem ~ Pls let us know if everything works fine with this version on you side.

So far everything is fine with studio driver 546.33.

Okay, that's great ! 

no__name

 

I still don't think it is a shadow, it is a lack of light being cast in that direction. If you want all-round illumination you need a different configuration of sources - try a cube, with only one division that would be only six fces (and it could be single-sdied).

Hm, done quick test again this morning and still not working.

Single faced/sided mesh (rectangle, disc and so on) or spotlights (as you can give advanced node to spotlight now) give the same behavior : it casts shadow (or shadow-like) pattern in every environement I tested. Therefore nothing "ghost".

Again it's very possible I'm doing something wrong. In that case, if someone from Daz could provide a small example would be neat. But so far it's not working.

barbult

Thank you, Daz developers for your hard work and the frequent beta updates this month!

Richard Haseltine

no__name said:

 

I still don't think it is a shadow, it is a lack of light being cast in that direction. If you want all-round illumination you need a different configuration of sources - try a cube, with only one division that would be only six fces (and it could be single-sdied).

Hm, done quick test again this morning and still not working.

Single faced/sided mesh (rectangle, disc and so on) or spotlights (as you can give advanced node to spotlight now) give the same behavior : it casts shadow (or shadow-like) pattern in every environement I tested. Therefore nothing "ghost".

What shape had you given the spotlight? The others would still be planar, and would have a dark area between the areas lit by the opposing sides just as the plane woulkd by my reasoning.

Again it's very possible I'm doing something wrong. In that case, if someone from Daz could provide a small example would be neat. But so far it's not working.

Dolce Saito

That shutdown optimization... I generally load/delete lots of hd characters on the scene, and because of this, that.. optimization... was... awesome! That hit the spot. Whoever put that on the backlog, have my greatest appreciation.

Most of my shutdown times are cut down between 50 to 90 percent.

(My inner thoughs; Toss the dforce code to that guy who improved this and we will have houdini :P )

barbult

Yes, this new fast shutdown is great for productivity.

crosswind

The way of closing has been changed in the latest version. The action of "Clearing Scene" that used to be a background job now turns into a "front-end job" (even if you save the scene first then Close or Ctrl + Q). Then the DS session in Task Manager will be fully closed in appr. 10-15 seconds.

As per the time that I counted for a couple of tests, the total time of closing is a bit shorter than before. However, TBH, I personally still prefer to the "background job" of closing... because it's a weird thing that the appplication window is still there after clicking Close even if I know what the application's doing...

johndoe_36eb90b0

@Richard Something is definitely wrong with Iray Ghos Light Factor advanced node.

First, setting it to 1.0000001 already has the same effect as setting it to 2.0.

Second, entering numbers 1.0 and 1.1 repeatedly sometimes does not reflect in the scene (Iray preview) until you click the + or - or move slider with mouse.

Third, having a mirror in the scene and a G8 character in front of it when you set IGLF to 2.0 the light source (disc geometry for spotlight, size 100) becomes invisible in the mirror reflection, but it is still visible in the character sclera despite sclera having a lower refraction index than the mirror surface. To eliminate sclera reflection you need to crank IGLF to ~11 which totally kills all mirror reflections.

crosswind

johndoe_36eb90b0 said:

@Richard Something is definitely wrong with Iray Ghos Light Factor advanced node.

First, setting it to 1.0000001 already has the same effect as setting it to 2.0.

Second, entering numbers 1.0 and 1.1 repeatedly sometimes does not reflect in the scene (Iray preview) until you click the + or - or move slider with mouse.

Third, having a mirror in the scene and a G8 character in front of it when you set IGLF to 2.0 the light source (disc geometry for spotlight, size 100) becomes invisible in the mirror reflection, but it is still visible in the character sclera despite sclera having a lower refraction index than the mirror surface. To eliminate sclera reflection you need to crank IGLF to ~11 which totally kills all mirror reflections.

As for the 1st point, according to the instruction, IGLF needs to be > 1.0, so any value > 1.0 will make it work.  2 is just the quickest way by manual entry... or clicking + sign will be also fine.

As for the 2nd point, haven't seen such a "laggy preview" when changing its value, maybe case by base...

As for the 3rd point, as an alternative, You may just zero Glossy Layered Weight and Top Coat Weight on the surfaces of Cornea and Sclera rather than cranking IGLF to a much higher value. Well, a mirror surface is different 'cause its Metallicity value has to be 1 and Metallicity layer overrides all the other layers in Iray Uber... 

trebor_gb_37ba5b8bb6

Hi

Does anybody know where/how I download 4.22.0.14, please?

DoctorJellybean

trebor_gb_37ba5b8bb6 said:

Hi

Does anybody know where/how I download 4.22.0.14, please?

If you already have a Beta version installed, then it should show up in IM. Just make sure that you have Public Beta selected in IM (Advanced Settings - Downloads).