|
|
Post by theropod on Nov 16, 2019 18:38:05 GMT 5
Yes, that only works if the images have sufficient resolution. You cannot scale a scalebar that you cannot even properly measure (with the scalebars in this picture, the margin of error is like 3 px in either direction, which means a 10% difference). But ideally, a scalebar should be in the hundreds of px long, and/or not have any blurry edges whatsoever (i.e. just a black rectangle that can be measured accurately).
|
|
|
|
Post by dinosauria101 on Nov 18, 2019 2:13:15 GMT 5
Alright, so here it is from Imgur: Palaeoloxodon namadicus (femur fragment) vs Mamenchisaurus sinocanadorum (poorly described but very complete skeleton), credit to Asier Larramendi and Greg Paul respectively for the skeletals. The elephant is scaled to 520 cm shoulder height and weighs around 22 tonnes, while M. sinocanadorum is scaled to 246 cm femur and would be about 75 tonnes. Scalebar is 1 meter.  |
|
|
|
Post by theropod on Nov 18, 2019 3:22:01 GMT 5
It’s impossible to measure the femur in this image with the accuracy to which you are giving the measurement (246 cm) because the femur isn’t just extremely blurry, but also only around 80 px long, so that alone would get this moved because your measurements aren’t reproducible. But we know that with a 246 cm femur the thing is supposed to be 35 m long in axial length, and if that elephant is 5.2 m (=156px) in shoulder height, the sauropod (1181px) is over 39 m long. So once more, 4 m off the mark.
|
|
|
|
Post by dinosauria101 on Nov 18, 2019 3:27:34 GMT 5
Drat! Not again.
I can try to rescale, but you'll have to pardon erroneous charts with very big animals as they are tricky to properly scale.
|
|
|
|
Post by theropod on Nov 18, 2019 3:34:01 GMT 5
They wouldn’t be tricky to scale if you only followed my very simple tutorial on the basics thread. I gave you everything you need to know to make a properly scaled comparison. I’m afraid I don’t know how to help you any more than that, but if you have a specific question be my guest, that might be better use of your time than posting one inaccurately scaled comparison after another in the hopes of getting it right by accident.
|
|
|
|
Post by creature386 on Nov 18, 2019 3:35:44 GMT 5
I think it's best to also give us some pixels in the future. The elephant's shoulder height was a bit odd, too. I have not calculated it exactly, but it was less than five times the scale bar.
|
|
|
|
Post by dinosauria101 on Nov 18, 2019 3:36:35 GMT 5
I can't download Gimp on my only device.
Oh well. I could practice pixel scaling so I can do it while making the charts.
|
|
|
|
Post by dinosauria101 on Nov 18, 2019 3:41:44 GMT 5
Alright, looks like I made ANOTHER scaling error - the elephant was supposed to be 500, not 520 cm.
Going by that scalebar, it looks to be ~485 cm, so it's still off but by not as much as we thought.
|
|
|
|
Post by theropod on Nov 18, 2019 3:45:14 GMT 5
The issue begins with the scale bar’s size being very imprecise. It’s so blurry it’s basically impossible to measure accurately, just like the femur length. What’s the length of this scale bar? The top bar? 30 px? or the bottom bar? 37 px? Or something in between? Am I supposed to guess? Based on the axial length, it should be about 34 px.  What’s worse, if this was the quality of image you were scaling from, I cannot help but think the only way you could have scaled it would have also basically been by guessing. A tiny little bit of blur may be ok if you have a scalebar that’s 200 px long and blurry by 1 or 2 px, but if it’s around 34 px long and blurry by 7, the imprecision just makes accurate scaling impossible. That’s why I measured the axial length, which is long enough a measurement (over 1000 px) that I can measure it relatively precisely. And if you scaled based on the femur or the scalebar at that terrible resolution, it’s no wonder the total length ended up being off by 10%. |
|
|
|
Post by theropod on Nov 18, 2019 3:46:11 GMT 5
I can't download Gimp on my only device. Oh well. I could practice pixel scaling so I can do it while making the charts. You don’t have a computer? What is "pixel scaling" supposed to be? |
|
|
|
Post by dinosauria101 on Nov 18, 2019 3:50:43 GMT 5
I can't download Gimp on my only device. Oh well. I could practice pixel scaling so I can do it while making the charts. You don’t have a computer? What is "pixel scaling" supposed to be? For some reason, there is no download option on the GIMP website when I click the download link. It says something along the lines of my device not being good for Gimp. I am going to try and use pixels instead of scalebars to measure, is what it means |
|
|
|
Post by theropod on Nov 18, 2019 3:52:44 GMT 5
Scalebars consist of pixels too, didn’t I just show that to you?
Well, what kind of device is it?
|
|
|
|
Post by dinosauria101 on Nov 18, 2019 3:55:44 GMT 5
What I meant by that was I ignore any pixels and only use scalebars. I'll try pixels too.
It's an incognito chromebook
|
|
|
|
Post by theropod on Nov 18, 2019 4:07:30 GMT 5
|
|
|
|
Post by dinosauria101 on Nov 18, 2019 4:47:45 GMT 5
Yeah I'll have to check those out.
What I meant by that was that I was not using the pixels to measure the animal's sizes relative to each other.
|
|
