NiranV Dean

Photoshopping is nice if you want to do product images and say get rid of backgrounds or want to edit in price, details etc, everything else can be done out of the box. Photoshopping in SL has reached similar levels to this common over-CGI-ing of movies. I'm not saying its bad, some things need to be edited (in movies) unless you want to blow up thousands and millions of dollars for a scene just to get a nice explosion. In SL however i see photoshopping as annoyance, i see no place for it (other than the aforementioned product information additions), i see no reason to ever edit a picture, SL's flaws are what makes it so unique, prims clipping into each other, texture artifacts, weird rigged mesh deformations, these are small flaws but they make SL what they are and finding creative workarounds for these inworld is much better than simply editing them out. That's the whole challenge i like so much. SL has a problem, find a way to fix it, there are ways, there are many ways in fact. To give you an example of this: Do you think this image looks nice? I don't. The whiteout trees, the alpha blending issues, the totally missing interaction of trees with shadows, the trees in the background being way too bright, there are too many issues in this pictures. Lucky for me i found ways to fix these issues. For this matter i added a feature to mask objects and fix what the creator didn't fix. Tune some settings, change things here, add slight depth of field, modify the color palette and poof: Same windlight all i did was play with all kinds of settings for a bit and fix the trees, completely different look. Here's another example of a nice place that suffers from the creators weird decisions in design: It looks nice, the whole scene has this cool warm, wet, swampy jungle atmosphere but the prim water is ruining the entire shot, double shadows, pixelated shadows, it doesn't reflect anything... for this scene i added a feature that allows users to change the water height locally, i used it to raise the water level while derendering the fake water prim, this changed the scene completely and forced me to make further changes to lighting, colors and the water as well. The end result is no comparison to the original shot and all that just by raising the water level, replacing the fake water with real water and tuning more settings to match the new water. This is simply something you can't do in photoshop, this wonderful magic of unlocking that last missing bit of potential in a picture and suddenly a whole new world of options open up, you'd be painstakingly hand editing this picture for hours, for what? When at the end you can still clearly tell that its photoshopped. This is also something you can't do in LL Viewer either. Nothing of the above final pictures is faked, all fake water has been removed, fake sunrays have been removed (if there were any), reflections are real (from water), the glow is real and coming from the viewer, the colors are unedited, shadows are real and depth of field is also not edited in. This is by no means something you can do by ticking two options and hitting snapshot of course, it still requires time (sometimes an hour or more) to fiddle with all the settings until you're happy with what you see but the outcome is absolutely worth it and it is something i can always show around and say "this is Second Life, how it ACTUALLY looks like", it's not a dream or an edited picture, its reality, in theory anyone can go to these places and fiddle with the settings until it looks "similarly" good (not saying the same because thats basically impossible unless you share the settings, trust me i tried) and if content creators saw the actual potential of their creations like in the above examples they could have used it... its quite sad how some really awesome places get dumbed down by seemingly unfixable stupid mistakes... but that's exactly what this is about, finding and fixing these with the tools available.

No, you're just getting slaughtered by the amount of small changes that have happened over the past 8 years. If you remap the keys exactly as in other Viewers (namely remap strafing back to Shift + A/D/Left/Right and remap rotating back to A/D/Left/Right, which are the major changes) you'll end up with exactly the same movement behavior as in all other Viewers, it's actually so easy you'll want to smash a desk to your face when you're done. There are many ways to do this but the simplest to do what i just described is: Double click the first entry 'Turn Right' , press D or Arrow Right (whichever you prefer), hit Bind. Double click the second entry 'Turn Left', press A or Arrow Left (whichever you prefer), hit Bind. Select the first two 'Move Left' entries (or just the one you want to replace), hit Delete button at the bottom. Select the first two 'Move Right' entries (or just the one you want to replace), hit the Delete button at the bottom. You can also just remove all Move Left/Right entries if you never use Shift + Movement Key and you can always just readd them later. You just change whatever you don't like and/or add whatever is missing to you, adding works pretty similar, you click 'Add', select an action, press a button and optionally one or more modifier keys and hit Bind. Pretty simple stuff. Not as simple as in games where you have a fixed amount of binds (usually a primary and secondary) and simply all available actions listed but i wanted the ability to bind as much as you want the way you want. ---- Preferences is obviously the biggest change from all Viewers because it needed something different, with options, options, options and more options being added i needed a way to present them logically and grouped together, while offering a consistent look that works across all preferences tabs that doesn't include just smacking more tabs into tabs into tabs into tabs until you're just swinging from tabs to tabs to find more tabs, besides if i made them all tabs you'd get a scrolling tab list which is ugly af and something i don't want, the alternative would be breaking options further down and adding tabs into tabs (in main tabs) which is just as ugly. A search option has been added (coming from Linden Labs, coming from Firestorm) that you can use to search for options, i've never used it for real yet but it should be good enough to at least give you a general direction where to look for whatever you're looking for if you just enter a keyword that is appropiate. ---- Jellydolls are a whole story for themselves. Complexity is calculated differently in BD due to complexity in its defaut state being too random and useless for the user. I've vastly changed the way objects are weighted compared to how LL does it and the result is a much more reliable albeit still 'roughly estimated' impact of someone's complexity with much higher values. BD in complexity can only be increased to a max of 800.000 before you have to either un-jellydoll people, have them as friends or simply set jellydolls to 'unlimited' because 800.000 is already considered a crazy high complexity (and most people, especially human avatars, easily go past that) which implies that someone is walking around with a decorated house worth of attachments on his avatar which will negatively impact your performance a lot, depending on the complexity sometimes just one person alone is enough to reduce your framerate by half, sometimes it takes 2 but its usually not far off of ~2 avatars. You can of course as said, simply set the complexity limits to 0 (infinite) and turn the feature off but unless you have at least some backup (like max avatars turned down to 5-10, to impostor people) you'll most likely see unusable framerates very quickly, but that's up to you then. Personally i'm confident enough to turn jellydolls off now that i have a Ryzen, i'm still getting around 30-50 FPS depending on the people that are currently fully rendered but i'd recommend not doing so unless you have FPS to spare. --- @Seicher Rae About entering contests, i had something similar happen with a Flickr group. It was a group about my Viewer for completely unedited pictures ONLY. When i looked through the gallery i saw lots and lots of clearly photoshopped pictures and i thought well time to add some unedited ones then, that's the rules aren't they? I added a few pictures to get them approved (yes they needed to be manually approved too) and shortly after i got kicked out of the group with the message that i violated the rules because my pictures are edited. Can you imagine? You, the creator of the very Viewer the entire group is about, getting kicked out of the group for violating the "No Edits" rule with unedited pictures. If i had any religion, i would have lost it then and there. Man i was so pissed.

That's the whole point. You DON'T do it in post. It looks bad. Also, doing it in post is only applicable for recorded footage, be it videos or pictures, you can't just do these in "post" unless you use third party injectors like Reshade and those will look bad too because Reshade doesn't have access to SL's depth buffer which it absolutely needs to do more than just some color correcting and fakish bloom. The whole point of the Viewer is that you have these things for real, in realtime, inside Second Life, not just as imaginary post processed fake show-off picture. You don't take Minecraft and photoshop volumetric lighting and clouds and shadows and all that in and say "WOW look, Minecraft looks so awesome" you go out download a shaderpack and use that to get these things for REAL in Minecraft and make it look actually nice. The reason some people (including me btw) are so snobby about their pictures is because they aren't photoshopped (if they are then they shall be excluded), i can take any of my pictures, the fanciest there is and say "This is how Second Life looks like", what can you? Show me a picture of your avatar that is photoshopped up and down, left and right and is basically a painted mona-lisa? Great but that's not how Second Life looks, thats just your imagination how you'd like it to look like. I don't see photoshopped snapshots as snapshots anymore, to me a snapshot, as fancy as it might be, is worthless if it was edited in post, its nothing more than worthless eyecandy that big companies have been shoving down our throats in media over the past 20 years, faked screenshots of a product that never existed, its simply a lie. Unlike some other people i hate being lied to and thats why i fight so much against photoshopped pictures, i want to end photoshopped pictures, i want to end the time we paint our never-to-be-realized imaginations into pictures and start a time where these can actually become a reality even if just with small steps, one at a time. Which brings me to pet peeve #2: Seeing ugly photoshopped pictures on the main page of Second Life when i could give them better pictures 90% of the time by just closing my eyes and randomly picking from any of my user's posted pictures. I've seen some absolutely beautiful and incredible pictures and instead we are presented with this... utter random photoshopped trash that looks like someone walked past a random location hit snapshot at a random point in time and called it a day, then touched it up some in photoshop and said "this is fine". It's absolutely demoralizing to see that photoshopped pictures are still so readily picked over much better unedited pictures, just as much as you saying "just add it in post". You are spitting into my face. According to you everything i've been fighting for the past 8 years has been a waste of time "when you can just add it in post". Thanks for nothing i guess.

I think what Alyona wanted to say here is that the general procedure taking a picture is the same across all Viewers, knowing the way around LL means you know the way around all Viewers. BD only appears to have a steeper learning curve on the surface because it surfaces (what a joke) more of the photography relevant options to the user. Almost all of the finetuning options for features like Depth of Field, SSAO and so on can be found in all Viewers, BD only adds new ones to these rarely and most of those were already present to some degree (such as shadow resolution, which is present in all Viewers but has seen decoupling from resolution and has more sliders to truly finetune it down to every single shadow map). While some effects are altered, they still work almost identical to how you'd expect them to work from other Viewers, the same options do the same change, it will just look slightly different. This 'slightly' different however is what makes the huge difference. BD lives from smaller details, slighter better shadows, slighter better depth of field, finer colors, smoother SSAO, subtle reflections on specular surfaces all these things are small if seen in an isolated case but alltogether they can make pictures truly shine in a way no other Viewer can. Those tiny things should be your main focus because (like Alyona implied) the baseline shot can be done in any Viewer. Pet peeve #1: People like this. I know i'm running into an open knife pointed at me here but what can i do, you basically summoned me. Your behavior right there is the very reason i behave so "atrocious". Maybe you should start looking at this from a different angle. Did i attack or insult you at any point? If so, ask yourself why, maybe you showed exactly this behavior before. If not, then ask yourself what the problem is, based on what you said you don't seem like someone who'd spend even 10 seconds wasting on a thought experiment why Niran might act this way. Take this very conversation as example. I'm here solely because you said this. I can guarantee you that if you started a conversation with me like that i'd definitely act exactly as you'd expect, absolutely atrocious. If you came to me at some point silently, inworld or discord asking me for help without ever mentioning that you said something like this (or that you are person X who said something like this somewhere), there would be no reason for me to be anything but nice and helpful towards you, unless of course you give me reason (someone actually did once).

Then you should know that BD has indeed features that you cannot achieve with the LL Viewer. I agree that anyone can take nice pictures with any Viewer, it's not the Viewer doing these shots, its the user after all. But you make it sound like you can make shots that look similar to what BD offers and i highly disagree, it's a simple fact that you cannot. BD offers unique features (i'm not talking about instagram like filters like greyscale and posterize), BD has a fully fledged highly accurate all-bone poser which would otherwise take you possibly hours to get a pose even remotely where you want it, in BD you can simply and easily "perfect" it and BD also offers better SSAO, Shadows, Depth of Field as well as its own unique graphic features like Screen Space Reflections and Volumetric Lighting which both can drastically alter the look of a picture. None of these are available in any other Viewer and none of their effects can be replicated. Here's an example of how those effects and the poser can be used in a scene:

Sounds like that's highly depending on how you implement this. You could make the Viewer save the loaded version instead of the full res version. Also if full res was always saved then i wonder why my texture cache hardly ever goes over 1GB even when i explicitely tell the Viewer to load everything full res and give me 8+GB of texture memory.

To be honest i'm quite unsure what part of the "constantly changing API" of either of those has any impact on how SL share works... i mean SL share is literally just sending a snapshot to their service with details like a description for instance. So unless the host/address changes constantly there's nothing that needs to "keep up" with their "changing API" or if they were to change the format these details/snapshots need to be send but those kind of things would happen with many years inbetween... On the other hand i don't care really. I've never used them and if anything i'd probably find them just as unreliable as pictures to profile which can sometimes randomly fail for no reason too. Generally speaking it is better to save a snapshot to disk and then upload it to any hoster anyway... for quality reasons.

14 scripts and 0.9MB is not much, sadly neither script count nor memory usage are worth jack *****. A single lightweight script can lag a SIM to near-death and potentially even crash it if we count fatal bugs. It really depends on what you do with said scripts. As stated here, you and/or others, it might be you, you and other people, or just other people. Generally speaking (and i know this will sound quite racist) but you're a human, your stuff isn't exactly "good", infact it is in rare cases only. Most human focused stuff is laggy and performance heavy because the content creators creating these don't know or don't care much, sadly, neither does most of SL's userbase. Everyone just complains but when you tell them they shut the doors and ignore you. Sadly as normal user you cannot find out whether it is your stuff or that of others or a combination of both, at least not without help. If you find one of these boards that show script count and memory usage, try those, most of them also show script time (besides memory usage that's the most important part), you want to keep your script time as tiny as possible.

Exactly what you said. You're lagging. Most likely because you and/or others are dragging the region down with heavily scripted items.

I like. We need more of this.

This is exactly what i meant. Any non-pocket calculator hardware will not benefit from turning off shaders or ALM, i see the same, turning off ALM reduces my framerate as well. But there's a reason for it too, given you have hardware with a decent hardware you basically shift some of the required cycles from CPU to GPU with shaders, you essentially free up extremely valuable CPU cycles which result in better framerates even though you are rendering "more".

Mac. *AUTISTIC SCREECHING* Self-built PC only. Ryzen 3600 @ ~200€ GTX 1060 6GB @ ~200€ 64gb 3200 DDR4 RAM @ ~150€ R470 Mainboard (for the Ryzen) @ ~150€ That's like 700€ total. Not exactly $55.000. This is not even anywhere near 1000. And that's pretty recent hardware, my previous hardware with an AMD FX 6200 and 16GB RAM and an old mainboard cost easily 300€ less and it ran SL fine for the past 7 years. A PC that can handle absolutely everything even outside SL doesn't even require 500€ and thats with pricing calculated from the time i bought them (7 years ago) today you basically get these for free. Seriously if people bought hardware rather than a stupid name that stood for inventing basic features that have been there for 15 years already they would get so much more bang for their buck.

I'd recommend setting everything beyond max. Pull up everything so high the Viewer keeps crashing, more crashes = less people = better for everyone. jk but seriously now, if you need to toggle even basic shaders and turn down trees, TREES. TREES! You might want to close the Viewer and do something in real life instead, or simply don't visit these events because if these settings honestly make a difference for you still then i wonder how you can even have these things on.

Well thats the thing, feed was going to die too, thats the reason they are bringing back the legacy profiles. If i remember correct it was said that a server update at some point will make us unable to post stuff to the feed and they dont want to fix this because it means a tremendous amount of work and bringing back legacy profiles is simply easier and less of a hassle than further supporting the feed. It was said on a TPV meeting before the legacy profile viewer was announced.

The profile feed is going to die anyway or do i remember this wrong? It's the reason they are going to switch back to clientside profiles... @Whirly Fizzle

I wish.

Not surprising. It's Catznip after all. I'm more suprised Firestorm allows setting more than 512 when Ansariel was clearly against raising texture memory limits.

Then i suppose i can't help you. I used the very same page to set up my build environment long ago when i first started making Viewers with Nirans Viewer, i didn't know any programming either nor did i even know what Visual Studio is or how any of the mentioned programs work, so i'd safely assume that anyone can set up a build environment using that guide.

Settings other than shadows (on/off, projectors) and shadow resolution, texture memory and draw distance have no real impact on your memory usage and disabling them or changing them wont fix this texture trashing issue. Texture memory "512mb" is actually more than that, its already 1gb minimum split into ~700mb + 300mb memory pools. Firestorm is the only Viewer aside of my own that i know that can set texture memory higher than the basic 512 (which is 1gb as mentioned before), i think FS allows up to 1 or 2gb even, that would be far beyond 4gb total.

The answer you are seeking is right here: I use this every time i have to set up my build environment.

It is also true that said XML file can be used to add new entries in said dropdown list because the Viewer reads and fills said dropdown from this list aside from the two default grids that are always present.

This is actually incorrect. Objects will almost always be fully rezzed regardless of whether they are occluded by something. The Viewer loads objects fully regardless of this option. If an object was rezzed, its "details" such as the resolution of its textures is solely dependent on the importance of said texture on your screen to the Viewer, you can summarize it as "the distance to your camera". Also object-object occlusion is kinda wonky and especially with houses and wonky bounding boxes of most items you should be super careful, object object occlusion can occlude stuff but usually does so in a very inefficient way, not only does it have a significant impact on your framerate which needs to be offset with render heavy objects being occluded to make up for the loss, its also quite bad at doing the only thing its supposed to do. To show you an example of how "effective" object object occlusion actually is: Off. On. I enabled the occlusion debug rendering for you to see what is being occluded, everything red is occluded by an object but not yet culled, everything that isn't visible anymore is culled. The difference between on/off is basically 1-2 items. Just because a tiny bit of one of the trees is poking out of the corner of the wall, you can imagine how this goes with buildings that have holes in them such as doors and windows.

There is still a way to connect to other grids. The Viewer contains a grids XML file in its app_settings folder, you can add grids there, although i dont know the formatting and required information to make this work.

CameraOffsetFrontView / FocusOffsetFrontView = Front View CameraOffsetGroupView / FocusOffsetGroupView = Group View ('shoulder' view) CameraOffsetRearView / FocusOffsetRearView = Rear View (default camera preset)

As Whirly suggested this is indeed an alpha fighting (z fighting) issue. Your friend's head is alpha and so is hair and his eyes. Unlike other Viewers Black Dragon does not make use of the pre-materials auto masking hack and instead shows you items as they are. There are several ways to fix this depending on how much control you have over said avatar and how much time you want to invest in fixing this issue. The Actual Fix: You replace all textures with ones that do not use alpha nor contain an alpha channel on export/save. You can do so by explicetly removing the alpha channel from a texture, in GIMP you'd go into Layers - Transparency - Remove Transparency Layer/Channel, then export the texture again and upload it, you can check whether you successfully removed the alpha channel by using the Local Texture feature and putting the texture onto a cube, if the cube shows the "Alpha Mode" dropdown in the Texture tab in the Tools window you did something wrong, "Alpha Mode" MUST be greyed, make sure you only select one face when doing so as selecting the entire object or multiple faces can grey out the dropdown depending on which face is counted as the first or last. No idea how to do this in Photoshop. Correct: Wrong: The Bandaid Fix: Select all faces/prims you want to "fix" and open the Tools window, select the Texture tab and choose "None" in the Alpha Mode dropdown if you dont need any alpha or select "Masking" if you alpha for things like eyebrows, don't forget to set a Cutoff if you choose "Masking" otherwise it will be just like "None". Doing this will tell the Viewer to handle the texture as a simple 1bit alpha thus practically making it a solid object, preventing these alpha sorting issues and improving performance too. From this: To this: The Pink-Glasses Workaround: Put on pink glasses again, ignore that the world is burning and re-enable Automatic Alpha Masks (Deferred). From now on you'll be jumping around happily in a huge flowerbed. --- Explanation: As you may have noticed in the pictures above, changing an object from "Blending" to "Masking" can have a big impact visually, both negatively as well as positively. In the case of my ears, they are made to use "Masking" mode hence, why they look no different using either mode, the difference between them are the many different visual bugs, starting with the pixelated shadows (can be seen on the picture at the tip), incorrect lighting, different material visuals, alpha flickering (z fighting, you are experiencing), not properly working post processing effects such as SSAO, Volumetric Lighting and as you already noticed Depth of Field completely ignoring the "depth" of said objects, higher performance footprint, in case of rigged attachments they cut out alpha objects behind them (seen on the pictures above, often happens to rigged hair on human avatars) and much more. In short alphas are bad and you want to not use them as much as possible and instead use "Masking" if you absolutely must use alpha, there are exceptions where "Blending" is necessary such as glass panes ofcourse. To Clarify: This issue happens on all Viewers if you toggle the auto masking hack off, however even though the hack does "hide" some of these cases, it is not perfect and utterly fails on harsher cases, a texture that is completely solid but contains an alpha channel is a surefire case to get catched by this hack, if you delete out a few pixels from the main channel, making it contain real alpha, this case would probably fall through as well and show you the same ugly issues. Also note that the hack does NOT fix all the issues nor does it a good job at hiding certain issues like pixelated shadows, pixelated shadows (especially on rigged attachments) will still appear even with the hack on ultimately requiring you to fix it manually. The hack is basically useless it just gives you as user the false impression that everything is fine when it is not.