The $10,000 10-inch challenge

[blonk]

Size estimates?
Natural Redhead Abby sees her first BBC

 

[blonk]

So far they've failed to prove any correlation between dick size and foot length, hand width, race, or any of the other supposed tell-tale signs. Yet, with posts like this one on this site, I can't help but think it tends to be inversely correlated with I.Q. I mean, come on man...

wait, wut
i asked her how many she's personally seen off-site, irl, etc. that are close or approaching the size in question. youre literaly retarded for thinking that question is in any way inappropriate.

 

[INCUBUS]

Has the 10" challenge become the new Porn link thread?
I mean what other reason are people linking these videos?
Not like any of these guys are a legit 10" and or who has ways of contacting them?
As the days goes by? This thread dies a little day by day.

 

[Lord_of_Goon]

Hey, just curious how tall are you?


@blonk Obviously a mans hand is gonna take up more space than a very small 5'2" Woman's hands which would make this same cock look fucking insanely crazy huge. By the way Julio claims 14" x 8". Why not believe his self declared size?

I’m 6’ on the dot.

 

[ItsAll4Kim]

Has the 10" challenge become the new Porn link thread?
I mean what other reason are people linking these videos?
Not like any of these guys are a legit 10" and or who has ways of contacting them?
As the days goes by? This thread dies a little day by day.

There's no reason for his thread to be kept alive. If someone wants to enter, they can.

Yeah, because no one was even close to be a real 10". They simply can't provide the proof as request.

If the goal was more realistic (or simply, the biggest), I'm sure there would be more serious contenders to discuss. It doesn't have to be 100% exact measurements (in case of volume, that would be impossible, as you said).

Don't you think something like this could be used to validate girth?
[video=medium]1918791[/video]
[video=medium]1353796[/video]

Idea: if you want to be more rigorous, take 3 measurements like this near the base, middle and head of the shaft and make a average. Then calculate the approximate volume using the length (hard ruler on top, etc etc). I don't know, just tossing ideas to make something fun around here. It's kinda of stale lately

It's a thread that's been running two years. That alone is miraculous. It's served far more than its original purpose.

Regarding your video; yes, that absolutely would work. And again, I'll remind you that not one guy has submitted a valid still photo. The one guy who came close to a valid measurement, who wasn't a contender, took two+ pages of dispute before submitting a 100% valid photo. Other than this, dozens of bullshitters. Expecting video is never gonna happen.

I can assure you that I have no interest in wading through a shitstorm of guys with seven inches trying to claim 8 or 9. It's tedious. Maybe someone else wants this, but I doubt it.

 

[Cum_is_Great]

Has the 10" challenge become the new Porn link thread?
I mean what other reason are people linking these videos?
Not like any of these guys are a legit 10" and or who has ways of contacting them?
As the days goes by? This thread dies a little day by day.

I only linked it to show proof of Julio Gomez claiming 14 x 8...

 

[Snakebyte]

in case of volume, that would be impossible, as you said

An exact measurement of volume is not impossible at all but prone to cheating.

 

[TinyPrincess]

I only linked it to show proof of Julio Gomez claiming 14 x 8...

He made the same claim on his old prison dating profile

http://www.theepenpal.com/find-a-penpal-2/show-ad/152/melvin-collins/25-35-male/

So if we use the good old porn inches to real inches conversion it would make him (14/1.66) approximately 8.4 real inches.

 

[Excelsus]

The proposals are to add girth and volume challenges, both of which are not possible to handle with digital media. A still photo will not show the back side of a tape wrapped around a penis.

True. However if the visible part of the tape is at least 2/5 of the measured girth, you can assume that the measurement is valid (as long as the dick is more or less round/oval shape).
For example: If somebody claims 6 inches girth (or 15.2 cm), at least 2.4 inches (or 6 cm) of the tape have to be visible. Of course this isn't 100% accurate, but it makes cheating quite a bit harder.

 

[hivelpsg]

Still on this?

Is there an analysis method or tool that is 100% objective, always correctly determines validity, and needs no interpretation of results?

Yeah latest computer vision algorithms, no human Interpretation needed. Thats something thats also valid and accepted as evidence at courts. Zero false positives. But thats nothing for hobbyists.

 

[g0nz0]

Yeah latest computer vision algorithms, no human Interpretation needed. Thats something thats also valid and accepted as evidence at courts. Zero false positives. But thats nothing for hobbyists.

That isn't exactly how these "latest computer vision" algorithms work. Interpretability is always needed... they are definitely mostly accurate but they can (and do) go wrong and make incorrect inferences.

When dealing with probability, it is unlikely to be the case that you can guarantee zero false positives - precision vs. recall is a trade-off.

 

[hivelpsg]

That isn't exactly how these "latest computer vision" algorithms work. Interpretability is always needed... they are definitely mostly accurate but they can (and do) go wrong and make incorrect inferences.

When dealing with probability, it is unlikely to be the case that you can guarantee zero false positives - precision vs. recall is a trade-off.

That was the case. Thats why i wrote "latest". The engine im using ( created by the Max Planck Institute for Informatics ) requires no human Interpretation and has no false positives. However thats nothing normal people have access to.

A different approach by Paul Wang is to detect everything photoshop or other editors can do. Its public available on github but experimantal. Also requires no human Interpretation.

 

[g0nz0]

That was the case. Thats why i wrote "latest". The engine im using ( created by the Max Planck Institute for Informatics ) requires no human Interpretation and has no false positives. However thats nothing normal people have access to.

Sorry, but this really does sound like gobbledygook. "Nothing normal people have access to" - I'm expecting "My name is Bond, James Bond" next...

Details: Any modern vision algorithm is either going to work based on classification (yes, this is a qualifying exemplar or not) or regression (measuring some quantitative aspect of it). The best approach we can achieve for either objective is to train the machine through examples, rather than prescribe rules. And we need lots of examples. Consequently, there will be errors (human labelling) in our training sets.

TL;DR: there is no magic bullet here that I am aware of, and I'm quite familiar with the state of the art. What you'll find in this domain is that algorithms are dime-a-dozen - the value is in the proprietary datasets. If there really is some magical new "engine" from the Max Planck Institut then I would greatly appreciate a link to an academic paper or announcement.

A different approach by Paul Wang is to detect everything photoshop or other editors can do. Its public available on github but experimantal. Also requires no human Interpretation.

By "Paul Wang", I presume you mean Sheng-Yu Wang et al's so-called FALDetector?

The FALDetector used convolutional networks and again large quantities of manually labelled training examples. Their system works for faces. Someone would need to construct a similar system for weiner, and it simply could not be made infallible.

AFAIK the FALDetector focused specifically on image warping. Certain techniques (digital airbrushing) are still very difficult to detect automatically. Furthermore, not every image manipulation is via PhotoShop - don't rule out optical effects (lenses), fake rulers/tapes, etc.

 

[hivelpsg]

Sorry, but this really does sound like gobbledygook. "Nothing normal people have access to" - I'm expecting "Bond, James Bond" next...

Details: Any modern vision algorithm is either going to work based on classification (yes, this is a qualifying exemplar or not) or regression (measuring some quantitative aspect of it). The best approach we can achieve for either objective is to train the machine through examples, rather than prescribe rules. And we need lots of examples. Consequently, there will be errors (human labelling) in our training sets.

TL;DR: there is no magic bullet here that I am aware of, and I'm quite familiar with the state of the art. What you'll find in this domain is that algorithms are dime-a-dozen - the value is in the proprietary datasets. If there really is some magical new "engine" from the Max Planck Institut then I would greatly appreciate a link to an academic paper or announcement.



By "Paul Wang", I presume you mean Sheng-Yu Wang et al's so-called FALDetector?

The FALDetector used convolutional networks and again large quantities of manually labelled training examples. Their system works for faces. Someone would need to construct a similar system for weiner, and it simply could not be made infallible.

AFAIK the FALDetector focused specifically on image warping. Certain techniques (digital airbrushing) are still very difficult to detect automatically. Furthermore, not every image manipulation is via PhotoShop - don't rule out optical effects (lenses), fake rulers/tapes, etc.

Its developed for governments and large firms, not for hobbyists. Not everything in this world is free and open source. Usually only crap is free.

Yes i was talking about FALDetector as a different approach. As i said its experimental and needs tons of more data to be reliable but it has potential for the future. I mentioned it because the question was do i know software that requires no human interpretation. And these two dont need it.

 

[g0nz0]

Its developed for governments and large firms, not for hobbyists. Not everything in this world is free and open source. Usually only crap is free.

I think I'll stop commenting after this, as your last post is just further nonsense, which is showing some severe misunderstandings of the technical subject matter at hand.

Work in deep computer vision techniques tends to be openly published. Companies like Google, Microsoft, Nvidia, Intel have contributed significantly to the collective knowledge. The "free" stuff is very far indeed from "crap" and due to the complexity of the stacks involved, there are no alternative closed stacks that outperform these open alternatives. They may be free in terms of cost, but certainly were not free in terms of the engineering hours invested in their development.

Yes i was talking about FALDetector as a different approach. As i said its experimental and needs tons of more data to be reliable but it has potential for the future.

If it needs "tons of more data", it will fall into the same problems w.r.t. human labelling and bias, and it cannot be a perfect system. It seems to me you do not understand the details of the subject matter you are espousing.

 

[hivelpsg]

I think I'll stop commenting after this, as your last post is just further nonsense, which is showing some severe misunderstandings of the technical subject matter at hand.

Work in deep computer vision techniques tends to be openly published. Companies like Google, Microsoft, Nvidia, Intel have contributed significantly to the collective knowledge. The "free" stuff is very far indeed from "crap" and due to the complexity of the stacks involved, there are no alternative closed stacks that outperform these open alternatives. They may be free in terms of cost, but certainly were not free in terms of the engineering hours invested in their development.



If it needs "tons of more data", it will fall into the same problems w.r.t. human labelling and bias, and it cannot be a perfect system. It seems to me you do not understand the details of the subject matter you are espousing.

So naive. Yes software sold for 1.2 million € to governments and large firms is just as good as the free ones. Sure.

The companies you mentioned only contribute what they dont need, outdated or easy to replicate. You are talking about for-profit companies. Doh.

 

[g0nz0]

So naive. Yes software sold for 1.2 million € to governments and large firms is just as good as the free ones. Sure.

The companies you mentioned only contribute what they dont need, outdated or easy to replicate. You are talking about for-profit companies. Doh.

Thank you so very much for your valuable contribution.

View media item 2895521
So, in summary to anyone catching up: as someone who works for one of the for-profit companies mentioned in the area of data science, it is my proposition that @hivelpsg is talking nonsense about a technical subject he is perhaps having some difficulty grasping, and @ItsAll4Kim is correct - there is no automated system available that could operate to the levels of accuracy he prescribed.

 

[Cm4hjl]

Had forgotten about this thread. Years gone by and no winner. Lots of talk, speculation, interpretation and opinions....... Bottom line, no 10 inch penis. The contest serves its purpose of limiting the BS, of which there is plenty. Guys will be guys. A little exaggeration lifts the ego and gets us laid. Ultimately, the “receiver” is probably more comfortable and satisfied with the smaller size anyway Yeah, I know......I just added another opinion.

 

[hivelpsg]

Thank you so very much for your valuable contribution.

View media item 2895521
So, in summary to anyone catching up: as someone who works for one of the for-profit companies mentioned in the area of data science, it is my proposition that @hivelpsg is talking nonsense about a technical subject he is perhaps having some difficulty grasping, and @ItsAll4Kim is correct - there is no automated system available that could operate to the levels of accuracy he prescribed.

As someone who actually owns one of these large firms holding government contracts i know very well what im talking about

I never said that its perfect. I said there are engines that require no human interpretation. I also never described the accuracy. I said there are no false positives due to the nature of how the engine works.

 

[g0nz0]

As someone who actually owns one of these large firms holding government contracts i know very well what im talking about

Of course you do. I wouldn't dream of suggesting otherwise

I never said that its perfect. I said there are engines that require no human interpretation. I also never described the accuracy. I said there are no false positives due to the nature of how the engine works.

Of course you didn't.

Neither do you understand the definition of accuracy for said systems, and how the false positive or false negative rate is how you determine accuracy...