J-20... The New Generation Fighter III

Status
Not open for further replies.

latenlazy

Brigadier
It is perfectly legitimate to compare two planes that are in side-profile. There are no camera tricks involved.

Except you have nothing objective to scale the two with. Assuming that two planes will have the same tire size or cockpit size is a huge assumption.

Just because you got a correct outcome with a faulty method does not mean the method is vindicated. Reasonably speaking the tire sizes are probably close, which would explain the result, but that again is still an unfounded assumption (even if it is a good one).
 
Last edited:

Martian

Senior Member
Except you have nothing objective to scale the two with. Assuming that two planes will have the same tire size or cockpit size is a huge assumption.

Did you look at the pixel-by-pixel analysis? The cockpit sizes were compared by Asymptote. The comparison was not meant to provide accuracy to three decimal places. It was intended to provide a best-approximation based on available evidence. The pixel-by-pixel analysis is certainly worth a lot more than your worthless speculations.
 

latenlazy

Brigadier
Did you look at the pixel-by-pixel analysis? The cockpit sizes were compared by Asymptote. The comparison was not meant to provide accuracy to three decimal places. It was intended to provide a best-approximation based on available evidence. The pixel-by-pixel analysis is certainly worth a lot more than your worthless speculations.

I did look at the pixel by pixel analysis, but it assumes that 1)each pixel represents the same amount of unit, and 2) There's no guarantee they have the same cockpit sizes. (Unless that information is published). These aren't necessarily wrong assumptions, but they are unfounded ones. Known objective references are necessary (for example a shared object between the two pictures, preferably with a known length).

Anyways, there is a problem with every estimate posted so far. I'm just griping at the assumptions being made.
 

Martian

Senior Member
I did look at the pixel by pixel analysis, but it assumes that 1)each pixel represents the same amount of unit, and 2) There's no guarantee they have the same cockpit sizes. (Unless that information is published). These aren't necessarily wrong assumptions, but they are unfounded ones. Known objective references are necessary (for example a shared object between the two pictures, preferably with a known length).

Anyways, there is a problem with every estimate posted so far. I'm just griping at the assumptions being made.

No. The assumption is that the cockpit sizes are approximate, which Asymptote clearly showed. The second point of reference was the close approximate size of the wheels. The third point of reference was the close sizes of the engines. The assumed level of confidence in the estimate was + or - 15%. It was not meant to be the definitive statement of the J-20's size. However, it did provide an accurate reference point and was better than the purely SPECULATIVE claims being made.

I am not happy at the cheap shots taken at me for providing the best approximation based on hard evidence. Go back and read those posts. Everyone else just threw a number out there. Mine was the closest approximation and it was accurate to within 1.6m.

I said the J-20 was roughly the size of the Su-27, which is 21.9m. Huzhigeng is stating the J-20 is 20.3m.

[(21.9m-20.3m)/(20.3m)]*100 = 7.88% error.

Who are you mindless chatterers to claim my insights were inappropriate?

Show me your old posts with a better estimate and with better evidence.
 
Last edited:

latenlazy

Brigadier
No. The assumption is that the cockpit sizes are approximate, which Asymptote clearly showed. The second point of reference was the close approximate size of the wheels. The third point of reference was the close sizes of the engines. The assumed level of confidence in the estimate was + or - 15%. It was not meant to be the definitive statement of the J-20's size. However, it did provide an accurate reference point and was better than the purely SPECULATIVE claims being made.
You can't show that two objects are approximate by using pixels. The pixels on one picture don't necessarily represent the same length as the pixels on another. For example, two pics of the same object with different pixel counts would have each pixel represent a different length. Similarly, you don't know that they use approximately the same sized wheels, or the same sized cockpit. While those are generally safe assumptions to make in this situation, they are nonetheless unfounded ones, which can make them problematic.

It's great that you admitted a margin of error, but that 10%-15% is a around a 2 meter +/- difference for an object of around 20 meters long. That could very well make the J-20 22-23 meters long (not that I think that's the case) depending on the error distribution.
I am not happy at the cheap shots taken at me for providing the best approximation based on hard evidence. Go back and read those posts. Everyone else just threw a number out there. Mine was the closest approximation and it was accurate to within 1.6m.
I apologize if you interpret my criticism as a cheap shot. I am only pointing out potential errors within the method, no offense intended.

I am also nitpicking here, but just because you got the right result doesn't mean the methodology was correct.
I said the J-20 was roughly the size of the Su-27, which is 21.9m. Huzhigeng is stating the J-20 is 20.3m.
[(21.9m-20.3m)/(20.3m)]*100 = 7.88% error.
Not to be a credit claimer but technically
latenlazy-albums-j-20-picture5392-j20j10comparison.jpg
this one was more accurate. People keep forgetting that one though.
Who are you mindless chatterers to claim my insights were inappropriate?

Not your insights. Your assumptions. Bad method is bad method, even if it turns out marginally accurate. While this time it might have been okay, it encourages the continuation of bad methodology which can create greater errors next time.
Show me your old posts with a better estimate and with better evidence.
I have several times actually...The tow truck estimates were smaller by about 1 meter, and the picture above practically nailed it. Anyways, no hard feelings?
 
Last edited:

Martian

Senior Member
You are inconsistent. You claim side-profile comparisons of two airplanes are baseless and yet, you use a picture-comparison to bolster your claim. Also, was that picture comparison originally made by you?

Furthermore, I provided a rough estimate that the J-20 was the size of a Su-27. After Asymptote conducted a detailed analysis and concluded it was smaller than a Su-27, it should have been obvious that the J-20 was about another meter smaller. I did not feel it was necessary to piggyback on Asymptote's post to make a refined estimate of 20.9m.

Between my post, insights, and Asymptote's pixel-by-pixel analysis, we came very close to the J-20's actual size. What is with the b.s. criticism that is being heaped on our hard work?

What the hell is this (see below)? You guys trolling me? I think I'll start putting your group of idiots on ignore.

You can't just assume they will have the same wheel size or cockpit size. Engine size might be a good starting point, but they're small enough so that a small variation of measurement can create large errors.

Comparing random pictures is the dumbest thing someone can do. Even Kim Jung Il can look taller than Clington (188cm/6 ft 2) if the angle is right.

Except you have nothing objective to scale the two with. Assuming that two planes will have the same tire size or cockpit size is a huge assumption.

Just because you got a correct outcome with a faulty method does not mean the method is vindicated. Reasonably speaking the tire sizes are probably close, which would explain the result, but that again is still an unfounded assumption (even if it is a good one).
 
Last edited:

latenlazy

Brigadier
You are inconsistent. You claim side-profile comparisons of two airplanes are baseless and yet, you use a picture-comparison to bolster your claim. Also, was that picture comparison originally made by you?
I did not claim a side profile comparison was baseless. I said assuming pixels of two pictures represented the same lengths, and wheels and cockpits of two planes representing the same size was baseless.

In the case of the pixels, if you had two pictures of the exact same object, but one had a smaller pixel count of the other, one pixel in the smaller picture would not represent the same length as one pixel in the bigger picture.

Pictures are a fine way to estimate when you have an objective reference point, something that can stay constant. I cannot tell whether Person A is taller than Person B if I just had a picture of each. If however, I had a picture of Person A standing with Yao Ming and another of Person B with Yao Ming, I can tell relatively who is taller. Then if I know Yao Ming's height, I can find the heights of Person A and Person B using ratios. It seems like you and asymptote tried to do that using cockpit sizes and wheel sizes. However, the problem is we don't know if the Su-27 uses the same sized cockpit and same sized wheel. That is where the bad assumptions come in. How do you know the J-20's cockpit size isn't 30% bigger, or the Su-27's wheel is 20% smaller? We can intuit that it isn't, but we can never be sure.

The picture I just posted overcomes that problem, because it's the exact same truck parked in front of the J-10 and the J-20. Even if we don't know the length of the truck, we can use the J-10's length, which we do know, to find the truck's length. Then we can use the truck's length to find the J-20's. Now, there is a question of whether perspective difference and the distance and angle difference between where the truck is parked in each is negligible. That would be a potential source of error, but at least we have one fixed variable, whereas a comparison between the Su-27 and J-20 side views have no shared fixed variables (Unless we know they shared a part, like the cockpit or the wheel--ironically the method you and asymptote used would have worked to find the YF-23's length, because we know it used the F-15's cockpit and wheel).

I do not claim credit for that picture. It came out in the first week of the J-20's revelation from another bbs, and Blitzo, I, and other forum members used it to get an estimate of the J-20's length, aware that it might not be completely accurate due to the aforementioned errors. That this picture wasn't commonly distributed has been unfortunate.
[/QUOTE]

Furthermore, I provided a rough estimate that the J-20 was the size of a Su-27. After Asymptote conducted a detailed analysis and concluded it was smaller than a Su-27, it should have been obvious that the J-20 was about another meter smaller. I did not feel it was necessary to piggyback on Asymptote's post to make a refined estimate of 20.9m.
Again, I apologize for being picky, but the criticisms were more about methodology than result. I do not doubt many people presumed it was around the size of a Su-27, but regardless of the result methodology was just a bad one.
Between my post, insights, and Asymptote's pixel-by-pixel analysis, we came very close to the J-20's actual size. What is with the b.s. criticism that is being heaped on our hard work?
Getting the right results does not mean using the right methodology.

For example, if I did a general comparison of the F-18 and F-15 using cockpit length, I would have to assume they were around the same length, when the F-15's cockpit is simply just more spacious than the F-18's. There was no hard reason for us to believe the Su-27 and J-20 had the same size cockpit even though we can intuitively guess that would be the case. Just going off that assumption to make a length estimate is basically a shot in the dark.

What the hell is this (see below)? You guys trolling me? I think I'll start putting your group of idiots on ignore.
Nope, just grading your homework. If you'd like to explain how you know for a fact the Su-27's cockpit and wheel sizes are the same as the J-20's please go for it. You might have caught onto something other people have overlooked.
 
Last edited:
I did look at the pixel by pixel analysis, but it assumes that 1)each pixel represents the same amount of unit, and 2) There's no guarantee they have the same cockpit sizes. (Unless that information is published). These aren't necessarily wrong assumptions, but they are unfounded ones. Known objective references are necessary (for example a shared object between the two pictures, preferably with a known length).

Anyways, there is a problem with every estimate posted so far. I'm just griping at the assumptions being made.

I'm more worried about the angles taken from the photos. As these were taken by fans, this meant their location and positions aren't fixed, and therefore can be anywhere and varying. Even if they did take it from the same angles, now the burden lies on the J-20. It can also be anywhere and not necessarily the same spot for both photos. It may not be at the same place at it was yesterday. Even if it is the same location that it parked, there are also no guarantees if it was slightly closer to the camera today than it was yesterday.
Lastly, the same goes to the truck. it can be slightly closer to the cockpit today than yesterday, and all that is sufficient to cause misreadings of distances between the shots, the angles, and the sizes of the objects.
Unless everything else in the environment were held constant, it is no controlled setting, therefore it's much harder to compare, and no guarantees of what we'll receive.
 
Last edited:

LesAdieux

Junior Member
japan plans to purchase the design of YF-23 from northrop as its 4G fighter

I watched the CCTV asian security program a couple of days ago, one of the guests, a major general of PLAA said japan might purchase the design of the YF-23 and used it to develop its next generation fighter.

japan is not allowed to buy the F-22 and it deems the F-35 not good enough, the debut of the J-20 has prompted some japanese to explore the possibility of reviving the YF-23.

I doubt it's possible to restart the YF-23 today, it was terminated in 1991, even northrop doesn't really have the print because the program was cancelled in the rather early stage.
 
Status
Not open for further replies.
Top