Are you sure AI10's gradients are superior to XX's?

[roman]

Here is the link to the page with some tests I made. Please follow the instructions in order to see the images properly.

Also get the XAR file to see how can you make XARA X gradients without visible banding.

Check the gradient tests here
XaraX & AI10 gradients

Also please keep in mind that I've set up the page only temporarily.

[roman]

Here is the link to the page with some tests I made. Please follow the instructions in order to see the images properly.

Also get the XAR file to see how can you make XARA X gradients without visible banding.

Check the gradient tests here
XaraX & AI10 gradients

Also please keep in mind that I've set up the page only temporarily.

[gwpriester]

Roman

I was looking at your examples and I do see some banding on the Xara images as well as the Illustrator images.

I wonder if this might have something to do with video drivers and how they redraw these gradients?

My system is three years old and I think my video card only has 16MB of video RAM.

Gary

Gary Priester

Moderator Person

<a href="http://www.gwpriester.com">
www.gwpriester.com </a>


XaraXone


http://www.gwpriester.com/flag.gif

[Klaus Nordby]

Dmagician, your samples are not very enlightening. My tests show conclusively that, alas, XX is troubled with banding in multicolor gradients wheras AI10 is not. If I feel like it, maybe I'll post some more examples - but why should I? I've already proved this irrefutably in my initial posting.

And there's really no such thing as "32 bit screen color" - this setting neither give you any more colors than 24 bit, nor does it give better CMYK colors (which are 32 bit, but not in ANY sense which is relevant to graphics cards).

K
www.xaraxone.com/FeaturedArt/kn/
www.klausnordby.com/xara

[roman]

First and most important thing. In my honest opinion, whole debate is pointless, because there is no winner in practical daily use you will experience flaws in both packages. But, if you want to prove your opinionion, Klaus, I think you should post more complex tests than that one in your initial thread.

---------------

I did these tests using following configuration:
P3733 256Mb RAM 16MB Gfx Ram/ Windows 2000, TrueColor (32 bit) color resolution.
The programs were put side by side Xara on the left AI10 on the right.

I've stretched the gradient from one corner of the square to another and played with intermediate colors to produce the same results in both packages.

Best is to check the screenshots in ACDSEE, explorer will "redither" the gradients.

Here is the example you maybe will find more interesting.

[roman]

.

[Klaus Nordby]

Dmagician, you're right! Totally strange - the "repeating" setting for the gradient removes the banding. An oddity indeed - can anyone at Xara explain this? Alas, its also an oddity of limited use: it will make gradients, well, repeat - and so there are many situations where it cannot be used. But for those where it can, I thank you, Dmagacian, for this weird discovery!

Nevertheless, this anomaly aside, my quality comparison between AI10 and XX stands. Also, in AI10 one can also change the midpoints ("Profile") in a multicolor fill - but Profiling is disabled in XX for multicolor fills, only working for two-color gradients.

K
www.xaraxone.com/FeaturedArt/kn/
www.klausnordby.com/xara

[roman]

I promise it's not the last weird discovery from me.

As far as AI & XX is concerned, I still think it's not worth discussion. They both have apparent flaws in their implementation of gradients. And they both have few things which other product lacks. Let's turn the page...

[Klaus Nordby]

Dmagician: "I promise it's not the last weird discovery from me."

Looking forward!

OK.

K
www.xaraxone.com/FeaturedArt/kn/
www.klausnordby.com/xara

[roman]

It's not the first... and it's not the last ;-)

I'm especially proud of The quickest way how to make metallic effects in XaraX

Roman