Exiled_Vizir

So I finish my tests, results are very interresting: - Sentence of 21 words, voice with 28 seconds of total sample duration, pure cpu PyTorch: 270 minutes - Sentence of 21 words, voice with 28 seconds of total sample duration, torch-directml/rocm-hip-sdk5.6.1 "I don't know what I'm doing" PyTorch: 35 minutes - Sentence of 21 words, voice with 43 seconds of total sample duration, torch-directml/rocm-hip-sdk5.6.1 "I don't know what I'm doing" PyTorch: 32 minutes In this test the new install is between 7 and 8 times faster, and better in the pure cpu mode going from 5 to 21 words the computation time is multiply by ~7 (36 minutes to 270) but for the torch-directml version only by ~3 (12 to 35 minutes). I've also did a test with the same sentence and 43 seconds of R De Niro sample and the duration is about the same (youtube lighter than netflix^^?). Now the strange things I noticed: - The CPU usage is in the range of 50-60% total CPU time for every tests - I tried to monitor my GPU during torch-directml tests: it did nothing or the monitoring software I used where unable to see any activity (by the way if you know a good software for monitoring a GPU doing computing tell me I will test it). To sum up, here are my opinions on these gains: - My previous installation is totally broken, so using the same CPU power it's way slower (it's the most likely) - My GPU had doing some works and I didn't see it (very possible also) - The recent version of tortoise have been heavily optimized - Linux kernel is way faster than NT kernel for this kind of computation (very unlikely). I've seen 10-15% better for linux for heavy compute workload at my work, some pathological case with a factor 2, but never factor 7, and in my case linux was virtualized.

Hi, I don't know if there are some others people whom play with tortoise-tts lately? My last extensive usage was for my video Grand Theft Arcanum and I used it in CPU mode for PyTorch. After that my conclusions was the tool is powerfull (maybe the most powerful open source TTS) but in CPU it's slow. I had some ideas to do with tortoise-tts but it will involve some extended testing and lot of iterations and it would be too long via cpu... But I see that recently AMD has made some progress on its GPGPU SDK, so I try to install it and try tortoise using a PyTorch accelerated by my Radeon. There is a mainly 2 ways to get PyTorch accelerated by RocM (the AMD GPGPU SDK): Native Windows (kind of beta state and seems very complex to install) Native Linux (should work but I'm lazy and I don't want to set-up a linux right) so I cut corners and try WLS2 (linux subsystem for windows) and after setup a small linux and some trial and errors annoying loops I manage to get a pytorch running and normally accelerated by torch-directml on top of rocm-hip-sdk5.6.1. Aparently it is a wrapper for Cuda code to work on any GPU via... DirectX 12 (yeah on a simily linux VM, we live interresting time...). Since I already loose my time on it, I thought let's finish this shit and I installed Tortoise-TTS on my embeded linux and after some DIY it worked! But.. when I launch it I had this warning: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling I ran the test entirely, and I though ok there is no CUDA but it seems faster than I used it for my video. So I make some run on my old tortoise-tts and on the new, and good news the run I done so far are between 3 and 4 times faster on the new install. So to be sure, I start a bigger sentence on one of my voices I maked for my previous video. The sentence is 21 words long, it should take more than 4 hours to generate on the pure CPU version. After I will test on my new maybe-gpu accelerated version where I hope it will take between 25 minutes (optimistic forecast) and 2 hours (pesimistic forecast). I will post the results here, stay tuned if you're interrested.

Ok, I see the interest with your video, the main drawback I see with bodies morphes is that that base game doesn't know they exist, so you can have clipping problems with clothes, also I see some of my poses broken when I use a model with the xbody activated, but not all . Also it explain why you some change in shadow.

Hi, yes the noticeable thing is is the impact on the lightning and the associate shadow, on the slim it's logical because the shape and size of boobs change, so does the shadows. But on the tanked one the shadows change on boobs but to me the shape/size is globally the same, it's like an invisible cast shadow in the morph one but not in the classic one. How does the morph are integrated in game? It's in the body add-ons or is it something like a bump map or normal which is interpreted by Hook5?

Hi, for the Latona BB I don't see major differences, the only things I see are the shadows, the shoulders (the BB + Morph seems to be a very little slimer) and the tits (not the entire breasts, just the tits) which are more darker on the BB Morph, but I think it could be same on the BB classic with some tuning of the YUV of the tits textures. For Latona Slim there is a clear difference: the Morph one has a more detailed torso and the breasts are/seems bigger with also more darker tits. So to sum-up, for me Latona Slim Morph is clearly better than Latona Slim, but for the BB one I'm not I could differenciate the 2 without infos.

Hi, yes I think it could be a very good base, with a whiter skin, less curly hair and maybe some change in the lips, it can do the trick. @Trapp Did you share this model already, or just use this mod : on an existing model is sufficient?

Hi, for bj and moaning sound I find the Christina Fox voice very good : You can also check the others pornstar in the voice and sound section : Voices and Sounds - Klub Exile. Also the Faustus Moans series are good.

Hi, you have 3 cases: - If you use vanilla graphics (no hooks, no reshade, etc) you adjust some settings of a room in the customiser (normally it's the the last tab if I'm not mistaken) and dipending of the room you can set the moment of day (night, morning, etc) and adjust some others elements (in the yacht room you can to be near an island or in plain sea by example). - If you use Hook5 paid you can modify the using the F4 menu and go to Level option or modify the _level_definition.txt of the room if it exists. - if you Hook3, 4 or 5 free you have to modify the _level_definition.txt of the room or create it if it doesn't exist. For Hook5 (and I think also for the 3 and 4) they are also default lightning parameters for rooms and for the customiser in the main11.fx file. Look in the customizer_sunlight and default_room_environment.

Hi, in general the Hook5 objects rooms are very good and have great lightning (see the Bladerunner styles room of State808 by example) but they tend to be big and you have to get a good PC to run them. In classic room retexture the one I prefered is the Hook5 Laboratory by Nicockman but it's very dark so you to experienced with Hook5 because some time you will to add your own lightning or modify the existing. In custom rooms the Keidos Mercs Town is a must have if you like cyberpunk environment. For beginners their are Deluxe Cafe, living room green H5 and the hair salon retexture. They good looking and use daylight so itis easy to take good picture or videos in them. In any case more the lightning is good in a room, the more it's difficult to integrate your model in it.

Hi, great model, and I like the black and white esthetic and the placement of the light

Hi, great models and light. When I see the competition presentation I thought that hexd33 was a freshman... But he is not, so you have great chance to win.

Hi, good job, that's what I call a character intro!

Hi, thank you for your presentation of the soft and for your sample, I will try it when because you're right it can be very useful for adding speech in a video.

I remade the poll because it gone and I don't know why.

That's strange there are 2 voices for Jeanet but only one voter... Is the forum haunted?

Hi, I'm a fan of your previous Fantasy Dress (the dark red/purple serie. This new set looks promising.

Hi, before my fight against the almighty Sequencer I started to work on a small sequence in a Fantasy set-up (with Elf, Sorceress... and a Frankenstein's Creature Wanabe). So here we go for a sequence with one of my v7.5 model upgraded for vX and Hook5 full, the retexture of the Lab by Nickcockman, great looking Hook5 clothes, the Xbody to go with and 4 persons... What could go wrong with all that stuff in the sequencer? Here what I did until now, I will upload it when the download issues will be fixed. Update of 25/03/23: Jeanette has the lead in the pool for now, I will let it open for one or two weeks, then I'll go with the winner for my sequence. Also I said I would upload the sequence when the upload will be available again. Now it's the case and I make good progress but I haven't finish everything yet and I think it will be long. Tell me in comments if you prefer that I finish the video then I post it with the new content I use for it, or if you prefer I post some contents before I finish it. Update of 10/03/23: I make some progress with some new poses, and I also work on the apprentice and her new clothes. Because yes it was her previous clothes which slows a lot my game... Update: the first episode is live, you can check it here. Tell me what you think. The location: l It's a cosy lab, a little windy and not very bright but with a lot of facility for your various experiments, including a catacomb in case you have shorten the life of one of your expe... need storage. Lilith Esperandrax: This young Elf has been employed to steal a formula book of the sorceress Aaricia Black, but she seems to curious for her own good... You can see more of her in that album. The apprentice: Her model will be on Jeanette or on Natasha. I made some screenshots so you can compare them and vote for the one you like the most: I also start to make new poses for the sequence: null But it will take time because I have a lot of poses for mid-sized character and in this sequence I will use a max size man and 2 small women.

Ah damn! Since I vote everything is at 20%... If you need a good sabotage, call the vizir! I will respond after my lamps exposition and before capturing the Calif's Princess.