Automated script vs Realcumber

General
24

Get ready for a wall of text.

I understand that everyone wants the best quality, but to be honest, the market doesn’t work like that. Mass production and low cost have always won. To do some comparisons;

  • why do people shop at IKEA instead of going to a carpenter or a small quality furniture store? price and a lot more products to choose from.
  • why to people buy cheap cars like a Honda or Toyota instead of buying a BMW or Mercedes? price at the cost of performance.
  • why do people buy cheap clothes that you have to throw away within a year instead of buying quality that can be used much longer? price and short fashion cycles making people want to buy new stuff instead of using the same clothes a second or third season.

That’s how it works today in most western countries. People want low prices and a large selection. Quality, performance etc. are the trade-offs in most cases.

When it comes to scripts no one wants crap scripts. I certainly don’t want it. But as @doublevr asked (or if it was @handyalexander), how far are consumers ready to lower their standards to get decent scripts to a lower price and/or getting a larger amount of scripts to select from? CzechVR is obviously still good enough to get recommendations here at ES despite @realcumber not scripting for them anymore. At least I haven’t read any complaints other than it takes time until new releases get a script. If scripts become too bad as a result of automated scripting then people don’t want to pay for it and spend money elsewhere. That’s how capitalism and competition works.

Another aspect is that tech often start out rather bad. Some of you are old enough to remember early voice recognition like when you called a customer care center on the phone or used it in your car? People made fun of how bad the early attempts were, especially when it came to recognize different accents. Today you have excellent voice recognition in speakers, phones, TVs and so on and very few are complaining because the tech evolved.

Scripting is a small market with little to no venture capital (I assume) and only indirect research in AI/image recognition (I highly doubt that any university has explicitly funded image recognition of dicks, breasts etc to facilitate scripting :slight_smile: ). So initiatives from SLR/@doublevr and TheHandy/@handyalexander should be encouraged as long as they don’t become anti-consumer trying to create closed ecosystems and other bad things as a result of it.

So be positive about technology advancement. Hopefully scripting using AI will be good enough given time. Personally I believe that the first step is just automating the bulk of the scripting and then humans will fix the tricky parts. Who knows, at some point in the future AI scripting might become SaaS (Software as a Service) where you pay the price of a script today to get a custom video scripted using AI with good results. No need to choose from a catalog anymore. (we can always dream :slight_smile: )

Finally, regarding the risk of setting a low standard for scripts due to automated scripting as it is now;
Money talks. Don’t use services that produce bad scripts. We can all help raising awareness of better alternatives. When some one asks for a movie to watch on Netflix you don’t answer that “this movie is crap, don’t watch that”, you anser “this movie is good”. Do the same with studios and scripts.

Don’t focus on the bad, focus on the good.

10 Likes
25

That is dependent on there being ENOUGH of a market for them to continue. Otherwise, it’s tyranny of the majority: those who want a better quality product are forced to accept what the “masses” have dictated.

26

Ding! Ding! Ding!

This is the eventual endgame. For reference, see every market, ever.

1 Like
27

And those who want a cheaper product have to accept a tyranny of the minority who is willing to spend more on quality scripts. The world is not only black or white. Just look at eroscripts - you can get thousands of scripts for free, but people are still willing to pay for scripts of good quality. There is a place for everyone and any efforts in making the scripts more accessible to everyone should be encouraged, not killed at the origin.

3 Likes
28

AI will also be able to write ideal scenarios in the future. But for this he must be trained. And feed him hundreds, maybe thousands of good hand-written scripts.

1 Like
29

Why do people shop at those types of stores and buy those types of products? I don’t think it’s only for the reasons you enumerated.

Do you know how hard it is to find an artisan crafter for quality furniture, or a good tailor making custom fits, or a cobbler/shoe repair shop, or electronics repair, etc etc etc? These things were common in my youth, were reasonably affordable, and are increasingly rare now. Where they even exist, they are prohibitively expensive.

Why?

Because people who want cheap, easy, crap drove them out of business, and for the few that remain, they have to charge an exorbitantly high price to make the profession worthwhile. The dynamic you described, if it plays out in our little niche in the way it seems to be going, will lower the bar dramatically enough that we will have lots of cheap, crap scripts and very few good scripts which become too expensive for most. It’s not as simple as “there will always be hand-crafted scripts to purchase” because fewer people will be able to make a living at it, and those that do will need to raise the prices out of reach of most of us.

I think your post was well reasoned and well written, but I don’t agree that this is the right (or inevitable) path. I do agree that we should speak with our money, but the other thing capitalism produces in bulk is a selfish streak a mile wide, so I imagine people will buy the cheap crap for instant gratification instead, as they usually do.

1 Like
30
2 Likes
31

I can’t see that coming. 100% automation isn’t even possible. Also each script creator makes it with its own flavor

32

An idea for SLR

Multi Level Script options

Level 1. $1.99 Basic script that doesn’t capture every movement.
AI created with minimal human editing.

Level 2. $2.99 Intermediate script that captures more of the movements.
AI created with more human editing, speeds and lengths are better synced.

Level 3. $3.99 Pro script that captures every movement while maintaining perfect speed and length sync.
AI created script with extensive human editing.

I don’t do a lot of VR scripts, I find them to be too intense for my taste. I always end up editing them down to my liking. Think of it like picking toppings off a pizza. Currently SLR only sells one type of pizza, and it’s loaded up with toppings. What’s missing is options, there are plenty of customers who want just a simple pepperoni pizza, or even a plain ass cheese pizza. So why not offer those choices to your customers?
I recently purchased the script for
Wheel Of Fortune - Mary Jane Evans
I do believe it was made by @Realcumber and it’s definitely a level 3 pro script. And just like I expected, it’s too much for me and I’m currently making edits. Given the choice I would have gone for the level 2 script, and would be inclined purchase more scripts more often.
I think by offering this script level option you might appeal to more customers, and at the same time increase script output. VR videos could be released with level 1&2 scripts, and level 3’s released at a later date.

This also opens up marketing options.

-Out Now on SLR Mandy Muse sucks you dry!-
-Scrpits levels 1&2 available now, level 3 coming soon-
-Purchase level 1 or 2 by this date, and receive a discount for level 3 upgrade when it’s released!-

Or even membership plans based on access to certain script levels.

This was something I was thinking about while editing the wheel of fortune script, thought I would share it with you guys. I may not be a regular customer, but I do think what you guys are doing is great. Keep it up fellas :+1:

1 Like
33

@SlowScripter Do you mean that there should be three scripts (one for each tier/level) for each video or do you mean that some videos only get a level 1 script, some level 2 and some level 3?

While the idea is novel I think there are a couple of drawbacks depending on how you thought it should work:

  • If there are several script levels for each video then it will require even more scripting work to be done for each video, not less, raising the scripting (i.e. production) cost per video.
  • If some videos just get mostly automated scripting, while other videos get the full love of extensive manual scripting, then you might find that your favorite video only have a level 1 script while videos you are not interested in have level 3 scripts. Not fun if you pay for level 3 subscription.
1 Like
34

You bring up sound economic points here. I would also like to add that I feel that there is a shortage of good scripters. Call me “uncultured” but I like NA and other mainstream sites and I’ve dropped a decent amount of money to get videos scripted on the scriptpoolers discord. There are some requests that have been opened for months and almost up to a year with a fair amount of money pooled and yet they aren’t being taken up. I honestly would pledge $100+ each for a select number of NA vids but nobody wants to pick it up so I’ve given up for the time being to get scripts commissioned.

Whilst NA has produced scripts that are not the best of quality they at least provide some satisfaction for people like me. From this point of view I’d rather have something than nothing as long as it provides utility to me personally.

2 Likes
35

What I’m picturing is three scripts per video, thus giving the customers options. As for production I don’t know how they do things at the SLR office. So my thought process is based only on my own methods.

Quick break down

  1. Run the video through the motion tracker, go through and take out any off, quick, red line jerks. This would be the level 1 script.

  2. Go through the whole script and adjust the points for timing and length.
    This is the level 2 script

  3. My last step is to test the script and take notes along the way, then go back and make any final edits. This where things differ between me and the more talented scripters. You, @Realcumber @Husky , your scripts are way more detailed then mine. I have the force, but I am no Jedi.
    This is where level 3 scripts fall in.

So based on my method, levels 1&2 could be done in a day. It’s that level 3 that’s going to take much longer.

The other thing to consider is the endurance and sensitivity issue. I’ve come across this on the forum and experienced it myself. If you take a look at my portfolio you’ll notice that I have done some stop and go scripts lately. The endurance issue is what prompted me to make them in the first place. Spend 5 to 10 minutes getting the Handy and everything ready to go, then 2 minutes in it all cums to an end. This gets worse with VR scripts. First you add in the extra excitement that come with VR porn in its self, then add on a interactive toy with a level 3 script. Maybe top it off with a little herbal enlightenment. Forget about it… As far as endurance goes, some guys never stood a chance. I think offering these lesser detailed scripts would definitely appeal to those looking to last longer in a VR session. And would also lead to more repeat business.

2 Likes
36

Alot of great discussion here everyone

I think its really important to ensure people always have a good experience (when it comes to anything in life) but especially from a business paying perspective, and especially for first time users.

First impressions are very important and getting returning users to keep coming back is key to long term success

If users have a ‘bad’ first time experience with a new toy, it becomes really easy to simply write it off that tech and content as gimmicky. This happens with great technology that ends up disappearing all of the time.

I feel that we have so much content already available to everyone now (free and paid), that quality should really maintain a priority focus, especially in order to be able to stand out as more content saturates the market

Fortunately @doublevr fully understands the importance of that, and that high quality will always be something you can count on at SLR as long as I can help ensure it

New tools to help speed up scripting are more than welcome if they can truly do so, and as long as the speed tradeoff is worthwhile for the (currently) guaranteed quality drop

I don’t think we’re there yet with useful enough AI tools tbh, as its a bit of an effort and battle to still use and the resulting data output hasn’t shown reliable enough to justify the time invested using it

The time setting up/retrying/watching motion tracking, checking point timing, and the time needing to tweak depths, shift points, delete points, and then place new points, all are additional button presses/clicks that add up time, which before you know it, you could have made a script from scratch that has much longer lasting value and high quality appeal instead

Motion tracking / AI simply cant understand what to do in so many situations that are complex even for humans to think through, so it will still always take a decent amount of time to manually script those. Because of that reason alone, I cant see scripts anytime soon just magically being pumped out relying only on AI without being significantly worse quality

Even ‘quick’ checking a fully premade script takes a fair amount of time, as full length VR scripts are often always long, and there can be between 5000 to 20000+ points/strokes in a single scene.

Mess up on a few points here or there might be ok in the beginning, but how many immersion breaking points will be acceptable enough before you want to keep using or ever return to that script or get tired and yearn for something better out there?

VR is all about going to that next level of immersion isn’t it?

Of course who knows what future tools will be able to offer, but for now, this is why I personally still prefer to manually script from scratch every time

7 Likes
37

If I’m understanding you correctly it seems like you’re expecting ai based scripts to be less intense and I don’t think that’s the case. VR scripts are intesne because the actors really go to town on eachother, the ai tracking will maintain this speed just with less accuracy.

That said I do think setting the ai up to automatically script as much content as possible then selling access to those scripts at a dramatically reduced price is a great idea. Although I think a $2 price difference is too small for how big the jump in quality is.

Personally I have no interest in scripts of a much lower quality to those SLR are creating at the moment but I expect many others hold different opinions. I think another scripts premium tier may be a good idea, maybe half the price and unlocks streaming access to all ai generated scripts.

In a perfect world the production of lower tier ai generated scripts wouldn’t reduce the number of high quality scripts that are released. I also think allowing the community to vote on a few upcoming hand made scripts would be a good way to deal with the issue of great scenes having sub-pqr scripts

2 Likes
38

the better ending

39

That might work for the small community we have now, but imagine if Nexflix or HBO would do that :slight_smile:

40

True enough

41

I think that people need to start to see quality scripts also as a community effort. Feedback is still something rare here, whenever you see there is something to improve let the scripter know about it.

By calling scripts sub-par, even yours you get into negative mindset, I always think that as long as the scripter is happy with the outcome the script can be considered good. With some effort and hard work you can make the script better, sure, but any script is better than no script at all. I always thought that it would be nice if some day we could have a thread where scripters could exchange their expierence for different ways of scripting the same segment of the video, just so people who want to improve their scripts can get an inspiration from others. I really like the idea of different tiers of scripts for commercial use, but I also think we could use something similar for free scripts too, maybe some form of rating-only website or some sort of conoseuir system where certain members of the community could rate scripts and give feedback on a weekly/monthly basis.

I think that AI is definitely a way to go in improving the time needed for a script to make, right now it can be used with some success, mostly for cowgirl/reverse cowgirl segments in VR, but give it some time and it can massively reduce the effort you need to put into script. Go back 10 years back and try to remember how smartphones were back there and you will see when I’m going with it.

2 Likes
42

As someone who doesn’t make scripts I try never to leave comments that could be perceived as negative, if someone has gone to the time and effort to create a script then I feel all I can do is thank them, comment on nice flourishes or keep my mouth shut. It’d feel very rude to point out areas that could be improved on and I can imagine scenarios where new scripters’ posts receive lots of criticism (constructive or not) which discourages them.

I was only calling potential solely ai generated scripts sub par, wasn’t trying to belittle anyone’s work.

I think this would be a good idea only if scripters had to opt in before their scripts were rated/given constructive criticism, again it’d be good to avoid scenarios where people upload a script then receive nothing but negative feedback

43

I can’t speak about others, but I know that personally I’d be quite happy if someone took the time to tell me what they think could be improved in my scripts. Constructive criticism is a good thing.
Obviously if someone just say that my scripts suck I’d be sad (well actually I’d probably just don’t care :smiley:), but if you say that x or y could be improved, that’s useful.

3 Likes