Jump to content

Differences between tga , jpg...


lucho
 Share

Recommended Posts

Hi:

I have a question about quality renders, which are the visual quality differences between the different images formats (like tga, jpg,bmp,etc),i want to know which format has the best visual quality and why, this could help me to choose the correct format for my images.

Thanks a lot

Luis Calatayud

Link to comment
Share on other sites

TGA and TIF are lossless formats, meaning there is no compression involved and as a result, no loss in quality. These two formats can also contain alpha channels. Jpg's are OK for a small file size, and also offers compression, so you can make the image size smaller with a reduction in quality. PNG's are also good for small file sizes. I would stay away from BMP's altogether.

 

I prefer TGA's and TIF's, but some people like to use JPG's with no compression because you get an image that is comparable to a TIF's quality, but with much smaller files.

 

-Chad

Link to comment
Share on other sites

PNG is also lossless and supports Alpha, as well as 16 bits per pixel if you need. PNGs are also supported by web browsers (whereas TGAs and TIFFs aren't).

 

I tend to use PNGs only, and JPG for web transmission, since it's a compressible and small enough for fast transmission.

 

Alexander

Link to comment
Share on other sites

I see, the best are TIF and TGA, but to work in internet JPG is the best to use(compatibility), could someone show a image renderized at the different formats?, a graphic reference could be helpful.

Thanks a lot for your answers

Luis Calatayud

Architect :cool:

Link to comment
Share on other sites

Correct me if I'm wrong, but actually "working" with .jpg files is not wise, as every time you save the image the quality degrades because it's compressed every time you save. Your best bet is to save your rendered (unedited - straight from your rendering program) image as either a .tif or .tga file since both will carry the alpha channel with them and will not automatically loose some of its quality with the very first save compression. Work with your images in these "lossless" image formats, and then save a copy to whatever format is appropriate with what you're doing next.

 

For web, save it as .jpg or something small.

 

For print, size isn't an issue usually, so keep it as a .tif with LZW compression (don't think LZW actually messes too much, if at all, with the actual image quality)

 

For sending to clients, save as .pdf so you can setup the page layout and image resolution before you ever give it to the client. You'd be amazed at how many clients can't seem to figure out why they can't see the entire .jpg image you emailed them at print resolution. They actually can get offended you sent them an image that isn't big enough for their screen! Then they'll gripe that the low-res image doesn't print well!!

Link to comment
Share on other sites

the argument about jpeg vs tif/targa has been covered several times. i argue in favor of use jpeg with the highest quality available over tif/targa becuase the difference is not noticable to the human eye. i.m.o. the amount of space you save vs the unoticable difference is more than enough to use the jpeg when rendering, escpecially when rendering an animation.

 

here is my test, very unscientific, and probably useless. this is an uncompressed AVI file. one of the image is a tiff, the other a jpeg. the images flip back and forth. when the shape on top of the image changes, the image also changes. the white shape itself was never compressed, so don't look at that for clues.

 

zoom in as close as you want.

 

remember, the AVI is uncompressed, so the tif didn't lose any quality, and the jpeg will have the same quality as when i imported it. i set this up with photoshop and aftereffects.

 

http://www.phase22.com/temp/out.avi [ 8 meg avi file ]

 

...if anyone else wants to set up there own test, by all means do so. but remember i am talking tif vs highest quality jpeg.

 

 

.

Link to comment
Share on other sites

Basically, I saved an image I recently finished as a .jpg format at the absolute maximum setting Photoshop CS would allow me to do. Duplicated that image, and saved it as the "testsaved" copy. I repeatedly opened the testsaved copy, painted a small pixel around the border, then closed (and saved at the previous maximum setting) the image, then reopened, and repeated until I had painted pixels around the entire outside of the image. I know, I saved MANY times. However, this proves the point of image degredation with .jpg files, even at maximum settings. I know they create smaller file sizes, but I don't know why anyone would want the first copy of their rendered work to already be degraded, even if it's ever so slight.

 

http://www.adcad.com/temp/testoriginal.jpg

http://www.adcad.com/temp/testsaved.jpg

 

Eric

Link to comment
Share on other sites

I believe that the format of the future will be EXR. We used it extensively on "I, Robot" and it was amazing. It supports multi channel data. It supports HDRI floating point data, has amazing lossless compression including doing autocrop data. It is also completely open source.

 

The splutterfish guys have released a free version plugin for it for MAX and it is really good. Combine that with the photoshop plugin and you are set.

 

http://www.splutterfish.com/sf/sf_gen_page.php3?page=plugins

http://www.openexr.com/downloads.html

Link to comment
Share on other sites

Basically, I saved an image I recently finished as a .jpg format at the absolute maximum setting Photoshop CS would allow me to do. Duplicated that image, and saved it as the "testsaved" copy. I repeatedly opened the testsaved copy, painted a small pixel around the border, then closed (and saved at the previous maximum setting) the image, then reopened, and repeated until I had painted pixels around the entire outside of the image. I know, I saved MANY times. However, this proves the point of image degredation with .jpg files, even at maximum settings. I know they create smaller file sizes, but I don't know why anyone would want the first copy of their rendered work to already be degraded, even if it's ever so slight.

 

http://www.adcad.com/temp/testoriginal.jpg

http://www.adcad.com/temp/testsaved.jpg

 

Eric

I see your point, but i guess i should clear up my working method with jpeg's.

 

i create a texture map. i create it at the best quality i can, then i save as a jpeg out of photoshop with a 10 value. i never re-save it again.

 

renderings tht i am going to post process (which is almost all), i will sometimes save as a tif, sometimes as a jpeg, depending on what i am doing. while i am processing, i typically am saving as a psd. when i am done doing all of the processing, and the image is complete. i save a s a jpeg.

 

when i am rendering an animation. i save the frames as jpeg unless i need a alpha channel. typically i do my post processing of animations in aftereffects. afteffects reads the original jpeg everytime, applies my post process, then i compres my video.

 

what i am saying is i rarely save an image as a jpeg twice. ...i would never recomend that to someone. i do however beleive that jpegs have a place in the process. i am not on a sgi system that is lighting fast, with terrabytes of memeory. i need to have an equal balance of speed and quality. sometimes that speed finds itself in the form of a jpeg at the begginginof the process, sometimes it finds itself as the final output of the process.

 

i would welcome a new format that exceeds anythign on the market today, but i need my image files to be adaptable to many systems, and applications. tiff and jpeg serve are highly established at serving this purpose. windows and mac come with viewers for these files already built in, although windows is a little slow on tiff with alpha.

 

about 6 years ago i remember jpeg2000 being billed as the next big thing, where is that today. if a new standard is adopted the entire industry needs to embrace it otherwise it will only work on local systems and not globally.

Link to comment
Share on other sites

I believe that the format of the future will be EXR. We used it extensively on "I, Robot" and it was amazing. It supports multi channel data. It supports HDRI floating point data, has amazing lossless compression including doing autocrop data. It is also completely open source.

 

The splutterfish guys have released a free version plugin for it for MAX and it is really good. Combine that with the photoshop plugin and you are set.

 

http://www.splutterfish.com/sf/sf_gen_page.php3?page=plugins

http://www.openexr.com/downloads.html

i read the description, but it was kind of technical. i think i need to try it out to fully appreciate it. does the sputterfish version work only with brazil, or does it work through with any render engine?

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...