7.87
#...
·
1
like
|
|---|
|
When I look through the big wall at Astrobin, I find a lot of pictures with a resolution of more than 3000px per edge. I suppose that this is one reason why Astrobin sometimes reacts very slowly. For most of the pictures this high resolution makes no sense at all. On the contrary, most pictures do not look really good when viewed in full resolution and take a very long time to load. Therefore I would like if the resolution for uploading pictures was limited to a defined maximum size. In a second upload step it might then be possible to upload with higher resolutions for spezial cases, e.g. for panoramas. A maximum edge length of 3000px would be a good value in my opinion. I would be interested to know what you think about this. CS Hartmuth |
6.49
#...
·
2
likes
|
|---|
|
Well, it really depends on the setup of each astrophotographer : maybe lots of us have over sampled images, particularly if you take seeing limitations into consideration ? But lots of sensors are in the range 10 - 20Mpixels (so you easily get over 6000x3000 images, which wouldn’t pass your threshold), and sometimes it is justified. Not to mention people using drizzle when processing under sampled images. Or people using AI like Topaz to upscale and/or sharpen smaller / under resolved images (I don’t say it negatively : for art purposes those images are nice, though they are not anymore scientifically accurate). Both ways output higher sized images... So instead of limiting the size of uploaded images, why not just put a message of confirmation ? Like « Do you really need that profusion of pixels ? » and let people make smaller versions if they feel it wasn’t justified ? Or we could even go crazy and use the data they feed in to check if the size of the image is relevant given their setup, but I don’t think Salvatore would go for such a complex system 🤯😉 In short I would let people be responsible for what they publish and just make a popup to remind them to not go overkill with pixels... That being said, I find Astrobin quite responsive on my side. |
8.87
#...
·
1
like
|
|---|
|
Hello Hartmuth, I would have no problem with doing so, but then pixel/scale function is not correct of course when reducing my pic. In original my chip of the ASI 6200 has about 9600x6400 pixels. By the way AB is advertising with the download of large files when having a premium membership. The download time for a pic is of course depending on the graphic card and screen card of your device. CS, Christoph |
18.84
#...
·
2
likes
|
|---|
|
Hello all, as I also own an ASI6200, I would be at the Bin=3 size or I would have to re-scale them client sided. I agree, in many situations these huge images are oversampled, but as @Christoph Lichtblau said, the huge file size is an argument for premium accounts. But one minor thing: The graphics card has no impact on the speed. It is not challenged in rendering 2D images at all. The speed is primarily determined by you internet connection and the latency in special. Also the the backend performance of the AB site is a factor. The arcsec/pixel scale is not impacted since this data is calculated while plate solving the uploaded image. So if you limit the upload maximum scale it would be no problem, or you have simply to re-run the plate solving process after shrinking. But actually I see only little advantage to limit the size in that way, but some negative side effects: e.g. for big panoramas / mosaics. CS Rüdiger |
 #...
·
1
like
|
|---|
| @Hartmuth, the size (either MB or pixels) uploaded by the user has no measurable impact on the speed that AstroBin serves you pages. With the exception, of course, if you choose to view the "real" size image. The image is resized anyway to the displayed size, and it happens on a separate server as a background job, so it doesn't interfere with the CPU load of the main server that serves the web pages. |
6.08
#... |
|---|
|
I half to agree with Hartmuth on this. There is no reason to keep the image at full size when it is a simple fix by just reducing the image size before you save it as a jpeg. Reduce the image size and keep the file size the same if you want, it will have no effact on how you will see it on astrobin, I typically reduce my images 50% to 2000 x 2000 and save the file at about 750kb and the images look great. CS Bob |
|
18.84
#...
·
1
like
|
|---|
|
Hi Bob, your observation is basically due to the use of jpeg compression you use. But jpeg is a lossy data compression algorithm. This works well on images with hard contrasts, but when ever you have smooth color gradients jpeg produces artifacts (depending on the quality set). That is the reason why it is a much better idea to use lossless compression as PNG or TIFF container. Here the quality of your image is defined by the information contained in you image and not what the jpeg compression leaves over. CS Rüdiger |
11.07
#...
·
6
likes
|
|---|
| I certainly want to be able to see my images at full resolution. That's why I'm here. Too many other sites crush the resolution that I've spent a lot of money and time to achieve. |
|
7.87
#...
·
1
like
|
|---|
So instead of limiting the size of uploaded images, why not just put a message of confirmation ? Like « Do you really need that profusion of pixels ? » and let people make smaller versions if they feel it wasn’t justified ? For me this is a good idea! I forgot to write my most important motivation for opening this dialogue. At the moment it is only possible to show the images in the respective screen resolution (up to Full HD) or in the maximum resolution. What the best resolution to display the image can therefore only be controlled via the maximum uploaded resolution. Maybe it would also be feasible to set a resolution preferred by the user as an intermediate level. This would reduce the problem with long loading times and it would be possible to show the images in a "reasonable" resolution. I didn't want to prohibit uploading in full resolution, but I wanted to introduce a second upload for it, if the picture exceeds a defined size-limit. Or an alternative possibility to determine the resolution offered to the viewer. CS Hartmuth |
#...
·
4
likes
|
|---|
So instead of limiting the size of uploaded images, why not just put a message of confirmation ? Like « Do you really need that profusion of pixels ? » and let people make smaller versions if they feel it wasn’t justified ? Agreed, I will add this. Say, above 8092 x 8092? AstroBin already can't process images larger than 16536 x 16536 in any case. Hartmuth Kintzel: This is probably better solved by implementing a nice image viewer for pixel peepers, to have a nice zoom box as you hover the intermediate size. It's on my list! |
8.42
#...
·
2
likes
|
|---|
Salvatore Iovene: This would be very welcome - I uploaded my mosaic images to google photos because that at least has a nice pan and zoom function (esp on mobile!) and it's a pain AB doesn't do this. To get back to the OP: with my scope and camera I have 1.24"/px, which is about well-sampled for my skies (slightly under sampled on the best nights). So looking at 1:1 is needed to see the smallest details captured. At same time my ASI183 has about 20M pixels or 5496*3672. The real kicker is mosaics, I did one with 6 panels of the veil nebula which ends up to about 9k*9k. I do agree that many pictures don't look good at 1:1, but I do think this is because astrobin does't make it very easy to zoom to that level so many people just don't see it (and thus uploaders don't bother). If a picture just doesn't look good at 1:1 it should be resampled, but that is a personal choice of the uploader. |
|
7.87
#...
·
1
like
|
|---|
Salvatore Iovene: I am looking forward to it.  |
9.85
#...
·
2
likes
|
|---|
| @Hartmuth Kintzel Dear Hardmuth, I am sorry, but I definitely disagree. Only the full resolution of an image tells about the quality. There are many pictures looking nice at a medium scale but look much worse at full resolution. I think that the IOTD staff has some sort if informal agreement to look at every image at full resolution before it gets an award. Personally I just love to look at well processed picture in full resolution. Looking at details, fine structures, tiny dust lanes in galaxies - this is pure eye candy. Missing this would make AB much less valuable for me. Best regards Fritz |
4.72
#...
·
2
likes
|
|---|
| Please don't limit anything ! Astrobin is one of the few places where we can post large mosaics or results from IMX455/similar high megapixel sensors. |
#...
·
3
likes
|
|---|
Nadir Astro: Definitely not gonna limit anything! |
|
7.87
#...
·
1
like
|
|---|
|
This morning I saw the new viewer with "click & zoom" function for the first time. For me it is a great innovation compared to the previous options! Also the new possibility of switching the "Plat-solution-overlay" on and off in the medium resolution is great. Thank you very much Salvatore |
#...
·
1
like
|
|---|
| I released it last night and so far the reception has been good. I myself like I this feature a lot and I've been using it like crazy. I'm guilty of being a pixel-peeper 😂 |
|
18.84
#... |
|---|
|
Hi, I have experienced some problems here. Safari on iPad has an issue. Also Firefox on PC (no vertical scrolling is possible). Also the panning direction is right opposite to usual touchscreen behavior. that's very irritating when working with tablet. I will will club the information together and send it to @Salvatore Iovene . CS Rüdiger |
 #...
·
1
like
|
|---|
Ruediger: *** Type your reply here *** I have the same issue on the iPad using Firefox. |
2.71
#... |
|---|
|
Using Firefox here. The click and zoom works, but yesterday I also saw and used the plate solve overlay button. Today it is gone. Ron |
