PI X3 - your initial view points
I think that the participants in this thread have been talking at cross-purposes and not quite communicating, so my synopsis of the issue will either help clarify the issue or add to the confusion -- take your pick.
I think that the point to make is that everything that you see on your computer screen is a graphic -- there is nothing about text that differentiates it from the standpoint of the monitor screen when viewing a web page.
When a browser displays "HTML text", it internally renders the font -- with anti-aliasing -- and then displays the resultant graphic image on the screen. When using Photoimpact to create something like a button with text on it, PI does this same operation using its own internal logic, and we then hand the resultant graphic image directly to the browser in some sort of image format (.jpg, .gif, etc.).
So it would seem that PI should be able to do just as well as the browser at converting text-on-a-background to sharp, clear graphics. This is what the OP is saying that PI does not do well.
Of course, there are other reasons why the output graphic may not be as good as one directly rendered from HTML text by the browser: JPEG images are lossy, so some anti-aliasing effects might be lost. And GIF is particularly bad since it has only a 256-color pallette to work from.
This still does not refute the observation that PI-rendered text often looks fuzzy even when you're working on it within the application. I have also noticed this and generally just live with it, avoiding trying to generate small button images with PI.
I think that the point to make is that everything that you see on your computer screen is a graphic -- there is nothing about text that differentiates it from the standpoint of the monitor screen when viewing a web page.
When a browser displays "HTML text", it internally renders the font -- with anti-aliasing -- and then displays the resultant graphic image on the screen. When using Photoimpact to create something like a button with text on it, PI does this same operation using its own internal logic, and we then hand the resultant graphic image directly to the browser in some sort of image format (.jpg, .gif, etc.).
So it would seem that PI should be able to do just as well as the browser at converting text-on-a-background to sharp, clear graphics. This is what the OP is saying that PI does not do well.
Of course, there are other reasons why the output graphic may not be as good as one directly rendered from HTML text by the browser: JPEG images are lossy, so some anti-aliasing effects might be lost. And GIF is particularly bad since it has only a 256-color pallette to work from.
This still does not refute the observation that PI-rendered text often looks fuzzy even when you're working on it within the application. I have also noticed this and generally just live with it, avoiding trying to generate small button images with PI.
Thanks for your clear and concise post, Gordon, that just about sums it up in a nutshell.
The first time I realised there was a problem with PI's anti aliasing when applied to small text, was when I was looking at some of the custom text that is on the Ebuyer.com website - they had really clear, and extremely small text, on their shop 'link adverts' (I don't know what you call them, they advertise offers and sections of their site on the right hand side of their pages). I tried to copy the style of one of these in PI, because it looked really professional on the PI site, and it was just impossible to create anything like the clear, sharp text that Ebuyer had created.
Here's one of the images that I saved several months ago, it was originally saved as a JPG so the quality isn't as good as the original - notice the difference between the first word 'Print' (generated by PI) and the rest of the small text.

This is exactly the sort of small graphic I want to create - obviously it's possible, as Ebuyer have done it - I'm just wondering which program they used. I hope it's not Illustrator (¢G500, isn't it?)
That effect is what I have been talking about - it's just impossible to produce anything like that with PI's anti aliasing - the results look terrible.
The first time I realised there was a problem with PI's anti aliasing when applied to small text, was when I was looking at some of the custom text that is on the Ebuyer.com website - they had really clear, and extremely small text, on their shop 'link adverts' (I don't know what you call them, they advertise offers and sections of their site on the right hand side of their pages). I tried to copy the style of one of these in PI, because it looked really professional on the PI site, and it was just impossible to create anything like the clear, sharp text that Ebuyer had created.
Here's one of the images that I saved several months ago, it was originally saved as a JPG so the quality isn't as good as the original - notice the difference between the first word 'Print' (generated by PI) and the rest of the small text.

This is exactly the sort of small graphic I want to create - obviously it's possible, as Ebuyer have done it - I'm just wondering which program they used. I hope it's not Illustrator (¢G500, isn't it?)
That effect is what I have been talking about - it's just impossible to produce anything like that with PI's anti aliasing - the results look terrible.
Last edited by sisom on Tue Jan 22, 2008 2:30 am, edited 2 times in total.
This is incorrect. Browsers do not anti-alias text. Windows (or OSX) do this if the option is turned on. Currently mine is set not to (I don't use an LCD screen, so no Cleartype) and I don't have the normal anti-alias active, which actually looks similar to PhotoImpact's. Also, PhotoImpact has its own routines for anti-aliasing. There are many ways to do this, and after looking at many other graphic programs it seems the routines used to scale graphics and anti-alias them produces the same fuzzy appearance.So it would seem that PI should be able to do just as well as the browser at converting text-on-a-background to sharp, clear graphics.
It's can be a little quirky to use, but try Inkscape. It's free, and a strong vector editor. I did a quick text export and it seemed to do as well as the image above.I hope it's not Illustrator (¢G500, isn't it?)
A couple of font editors that might help with your issues, sisom.
http://www.blide.org/FONTMACHINE.htm
http://www.codehead.co.uk/cbfg/
http://www.blide.org/FONTMACHINE.htm
http://www.codehead.co.uk/cbfg/
Get back to the topic!
To get back to the topic - what about PI X3?
Well, I just upgraded - Corel let us wait for more than 3 months, and that is all they can give us.
Some minor improvements plus a few Corel programms no one needs.
It seems that the auto-process features did get better - it was barely usable in PI12, now you can give it a try.
There is a big regression with RAW import, at least my Nikon D80 files look totally crappy, whereas the import in PI12 is excellent.
Well, I just upgraded - Corel let us wait for more than 3 months, and that is all they can give us.
Some minor improvements plus a few Corel programms no one needs.
It seems that the auto-process features did get better - it was barely usable in PI12, now you can give it a try.
There is a big regression with RAW import, at least my Nikon D80 files look totally crappy, whereas the import in PI12 is excellent.
Whilst in Photo impact X3- Not yet impressed/page1dafalias wrote: ...There is a big regression with RAW import, at least my Nikon D80 files look totally crappy, whereas the import in PI12 is excellent...
So - is RAW buggy or notheinz-oz wrote:...It all depends what you use it for mainly. I like its improved ability to handle RAW images and the possibility to work with 48 bit RGB images. Than again, 48 bit was already possible in PI 12...
-
heinz-oz
-
John Moran_2
- Posts: 81
- Joined: Sun Mar 16, 2008 11:47 pm
Made the jump to PhotoImpact X3 (full boxed version) from PI 6 due to purchase of a Vista laptop. Pleased to see that the functionality of PI 6 is not only retained, but also greatly improved. On top of that, PI X3 seems pretty solid on Vista in preliminary mousing around. Until I got a DSLR with RAW file capability a year ago, I never felt a compulsion to either upgrade from PI 6, nor jump to another brand of editor. Corel's acquisition of the product also gave additional pause to contemplate.
In the past few days I have found that the included Corel MediaOne is something of a necessity for browsing, although I did "unplug" its intrusion somewhat. MediaOne, with its ability to embed Photo Info keywords and other metadata, instead of maintaining that data in a separate album or gallery database, as in PI 6 Album, is a welcome improvement.
I have not yet tested the Printer output from PI X3 on the Vista laptop to either my b/w K-M laser printer or my newly acquired Canon MP970 all-in-one color pinter. Printing from PI 6 was always a weak point in my view, and as another user has pointed out, Canon bundles some great software for that purpose anyway.
End of initial comments on PIX3. Begin comments on small fonts, as mentioned by other users on this thread:
•Unless small font comparisons are made on several different monitors of several different types and screen sizes and resolutions, by several different users of different ages and bifocals, with and without Windows anti-aliasing, no conclusions can be made. I was amazed at how much better all fonts looked when I quit using my generic 17 inch CRT a year ago.
•Yes, vector open fonts may tend to look better when generated directly, rather than being pasted in a from a bitmap file. But yes, there are better looking fonts out there than the standard Windows "Big Four". However, for those of us over 60 looking at small fonts on a CRT monitor, it really doesn't make any difference. We can't read any of them without copying them out to Irfanview and blowing them up.
In the past few days I have found that the included Corel MediaOne is something of a necessity for browsing, although I did "unplug" its intrusion somewhat. MediaOne, with its ability to embed Photo Info keywords and other metadata, instead of maintaining that data in a separate album or gallery database, as in PI 6 Album, is a welcome improvement.
I have not yet tested the Printer output from PI X3 on the Vista laptop to either my b/w K-M laser printer or my newly acquired Canon MP970 all-in-one color pinter. Printing from PI 6 was always a weak point in my view, and as another user has pointed out, Canon bundles some great software for that purpose anyway.
End of initial comments on PIX3. Begin comments on small fonts, as mentioned by other users on this thread:
•Unless small font comparisons are made on several different monitors of several different types and screen sizes and resolutions, by several different users of different ages and bifocals, with and without Windows anti-aliasing, no conclusions can be made. I was amazed at how much better all fonts looked when I quit using my generic 17 inch CRT a year ago.
•Yes, vector open fonts may tend to look better when generated directly, rather than being pasted in a from a bitmap file. But yes, there are better looking fonts out there than the standard Windows "Big Four". However, for those of us over 60 looking at small fonts on a CRT monitor, it really doesn't make any difference. We can't read any of them without copying them out to Irfanview and blowing them up.
John, the differences between the fonts are quite clear, I've done several blow ups of the problems in this thread, it makes no difference what monitor, resolution, etc. that you view them on, it's just down to the PI using a different method of anti-aliasing than Windows uses.
ps Reactor, thanks for the links to the two programs, but I couldn't work out how I would use them for what I wanted. Is there something obvious I'm missing?
ps Reactor, thanks for the links to the two programs, but I couldn't work out how I would use them for what I wanted. Is there something obvious I'm missing?
-
John Moran_2
- Posts: 81
- Joined: Sun Mar 16, 2008 11:47 pm
OK, I see your "point" now. It seems that PI X3 still has some work to do on small fonts, or it may be that PI Arial 10 is just a pixel shy of being tall enough. and Arial 10.5 or 11 is not available.sisom wrote:John, the differences between the fonts are quite clear, I've done several blow ups of the problems in this thread, it makes no difference what monitor, resolution, etc. that you view them on, it's just down to the PI using a different method of anti-aliasing than Windows uses.
Suggest you try the Corel Painter Essentials app included with PI X3 for your bitmapped text on a colored background base layer. If that is not satisfactory, please reconsider your opposition to using another application and try pasting a PrtSc of small text on colored background to PI X3 from another Text Editor with Draw capabilities, as suggested above initially.
Oh, I found a Draw module on my computer that works well, comes at the right price, and reportedly works on Linux, but cannot name it because I just signed onto the board and it is not a Corel drawing product. Good luck.
-
John Moran_2
- Posts: 81
- Joined: Sun Mar 16, 2008 11:47 pm
Did you try Corel Painter Essentials yet? It may already be on your computer, and seems to have good small font control! As to unmentionable open source OpenOffice.org Draw, (Not Inkscape) I had never found a use for Draw until last night while following up on your postings. Good luck, and thanks for giving me a reason to look at Painter Essentials.
